{ "scope": "Explicit path:line references in code and source headers only; not a complete bibliography or quote verifier", "checked_at_utc": "2026-09-08T02:51:38.855628+00:00", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" }, "checked_engines": [ "sglang", "vllm" ], "inventory_sha256": "7cd7b790b2c427d8663aae2f588f1c153e783f9dc079835de51bcce102fae6b1", "counts": { "range-exists": 3528 }, "resolution_counts": { "explicit-engine": 2941, "reviewed-context-and-pinned-symbol": 362, "same-page-qualified-anchor": 84, "same-section-qualified-anchor": 12, "unique-exact-path-in-pinned-trees": 129 }, "unresolved_by_path": {}, "pages": { "00-foundations/00-01-what-inference-is.html": { "range-exists": 30 }, "00-foundations/00-02-decode-loop-and-kv-cache.html": { "range-exists": 25 }, "00-foundations/00-03-gpu-architecture.html": { "range-exists": 29 }, "00-foundations/00-04-roofline-and-arithmetic-intensity.html": { "range-exists": 23 }, "00-foundations/00-05-numerics.html": { "range-exists": 27 }, "01-the-core-loop/01-01-prefill-vs-decode.html": { "range-exists": 22 }, "01-the-core-loop/01-02-latency-metrics-and-slos.html": { "range-exists": 46 }, "01-the-core-loop/01-03-continuous-batching.html": { "range-exists": 34 }, "01-the-core-loop/01-04-the-scheduler.html": { "range-exists": 29 }, "01-the-core-loop/01-05-chunked-prefill.html": { "range-exists": 38 }, "01-the-core-loop/01-06-pd-disaggregation.html": { "range-exists": 41 }, "02-memory-and-kv-cache/02-01-kv-cache-sizing.html": { "range-exists": 28 }, "02-memory-and-kv-cache/02-02-pagedattention.html": { "range-exists": 40 }, "02-memory-and-kv-cache/02-03-prefix-caching-hash.html": { "range-exists": 17 }, "02-memory-and-kv-cache/02-04-radixattention.html": { "range-exists": 43 }, "02-memory-and-kv-cache/02-05-kv-quantization.html": { "range-exists": 65 }, "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html": { "range-exists": 59 }, "03-attention-kernels/03-01-naive-attention-and-online-softmax.html": { "range-exists": 33 }, "03-attention-kernels/03-02-flashattention.html": { "range-exists": 27 }, "03-attention-kernels/03-03-decode-attention.html": { "range-exists": 38 }, "03-attention-kernels/03-04-flashinfer-and-backends.html": { "range-exists": 69 }, "03-attention-kernels/03-05-mha-mqa-gqa-mla.html": { "range-exists": 33 }, "03-attention-kernels/03-06-positional-encoding-and-long-context.html": { "range-exists": 49 }, "04-quantization/04-01-the-design-space.html": { "range-exists": 36 }, "04-quantization/04-02-gptq-awq-smoothquant.html": { "range-exists": 36 }, "04-quantization/04-03-formats-and-kernels.html": { "range-exists": 73 }, "04-quantization/04-04-when-quantization-pays.html": { "range-exists": 31 }, "05-parallelism/05-01-tensor-parallelism.html": { "range-exists": 42 }, "05-parallelism/05-02-pipeline-parallelism.html": { "range-exists": 36 }, "05-parallelism/05-03-data-and-expert-parallelism.html": { "range-exists": 55 }, "05-parallelism/05-04-collectives-and-topology.html": { "range-exists": 34 }, "05-parallelism/05-05-worker-orchestration.html": { "range-exists": 56 }, "06-decoding-algorithms/06-01-sampling-on-gpu.html": { "range-exists": 61 }, "06-decoding-algorithms/06-02-speculative-decoding.html": { "range-exists": 40 }, "06-decoding-algorithms/06-03-draft-sources.html": { "range-exists": 42 }, "06-decoding-algorithms/06-04-eagle-medusa-mtp.html": { "range-exists": 65 }, "06-decoding-algorithms/06-05-structured-decoding.html": { "range-exists": 71 }, "06-decoding-algorithms/06-06-speculative-decoding-zoo.html": { "range-exists": 99 }, "07-architectures/07-01-mixture-of-experts.html": { "range-exists": 53 }, "07-architectures/07-02-mla-in-depth.html": { "range-exists": 53 }, "07-architectures/07-03-ssm-and-hybrids.html": { "range-exists": 58 }, "07-architectures/07-04-multimodal-serving.html": { "range-exists": 72 }, "07-architectures/07-05-lora-and-pooling-models.html": { "range-exists": 63 }, "08-compilation-and-runtime/08-01-cuda-graphs.html": { "range-exists": 56 }, "08-compilation-and-runtime/08-02-torch-compile.html": { "range-exists": 51 }, "08-compilation-and-runtime/08-03-triton-and-cutlass.html": { "range-exists": 42 }, "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html": { "range-exists": 41 }, "09-serving-system/09-01-openai-api-surface.html": { "range-exists": 42 }, "09-serving-system/09-02-tokenization-and-detokenization.html": { "range-exists": 66 }, "09-serving-system/09-03-request-lifecycle.html": { "range-exists": 99 }, "09-serving-system/09-04-routers-and-load-balancing.html": { "range-exists": 50 }, "09-serving-system/09-05-observability-and-failure.html": { "range-exists": 48 }, "10-benchmarking/10-01-workload-characterization.html": { "range-exists": 20 }, "10-benchmarking/10-02-benchmark-harnesses.html": { "range-exists": 44 }, "10-benchmarking/10-03-honest-benchmarking.html": { "range-exists": 26 }, "10-benchmarking/10-04-batch-invariance.html": { "range-exists": 55 }, "10-benchmarking/10-05-profiling-and-capacity.html": { "range-exists": 33 }, "11-vllm-deep-dive/11-01-repo-map-and-v1.html": { "range-exists": 38 }, "11-vllm-deep-dive/11-02-engine-layers.html": { "range-exists": 66 }, "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html": { "range-exists": 40 }, "11-vllm-deep-dive/11-04-model-runner-and-backends.html": { "range-exists": 62 }, "11-vllm-deep-dive/11-05-vllm-extension-points.html": { "range-exists": 86 }, "12-sglang-deep-dive/12-01-repo-map-and-processes.html": { "range-exists": 38 }, "12-sglang-deep-dive/12-02-sglang-request-path.html": { "range-exists": 45 }, "12-sglang-deep-dive/12-03-radix-cache-in-code.html": { "range-exists": 108 }, "12-sglang-deep-dive/12-04-frontend-grammar-router.html": { "range-exists": 75 }, "13-comparison-and-frontier/13-01-head-to-head.html": { "range-exists": 56 }, "13-comparison-and-frontier/13-02-choosing-an-engine.html": { "range-exists": 31 }, "13-comparison-and-frontier/13-03-open-problems.html": { "range-exists": 39 }, "13-comparison-and-frontier/13-04-what-to-build-next.html": { "range-exists": 47 }, "index.html": { "range-exists": 1 }, "labs/01-measure-your-gpu-roofline/README.html": { "range-exists": 6 }, "labs/02-kv-cache-sizing/README.html": { "range-exists": 4 }, "labs/03-continuous-batching-effects/README.html": { "range-exists": 7 }, "labs/04-prefix-cache-hit-rate/README.html": { "range-exists": 28 }, "labs/05-chunked-prefill/README.html": { "range-exists": 10 }, "labs/06-quantization-tradeoff/README.html": { "range-exists": 25 }, "labs/07-tp-scaling/README.html": { "range-exists": 12 }, "labs/08-spec-decode-acceptance/README.html": { "range-exists": 16 }, "labs/09-structured-decoding-overhead/README.html": { "range-exists": 23 }, "labs/10-profile-a-decode-step/README.html": { "range-exists": 16 }, "labs/11-end-to-end-benchmark/README.html": { "range-exists": 18 }, "labs/12-read-a-request-through-the-code/README.html": { "range-exists": 23 }, "labs/13-batch-invariance/README.html": { "range-exists": 14 } }, "records": [ { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/metrics/loggers.py", "first": 283, "last": 285, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L283-L285" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 685, "last": 695, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L685-L695" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/config/cache.py", "first": 80, "last": 87, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L80-L87" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 559, "last": 562, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L559-L562" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 782, "last": 793, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L782-L793" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/config/scheduler.py", "first": 44, "last": 44, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L44-L44" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/engine/arg_utils.py", "first": 2580, "last": 2671, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2580-L2671" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/entrypoints/offline_utils.py", "first": 590, "last": 626, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/offline_utils.py#L590-L626" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/engine/async_llm.py", "first": 601, "last": 612, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L601-L612" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/engine/async_llm.py", "first": 616, "last": 621, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L616-L621" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/engine/llm_engine.py", "first": 105, "last": 111, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/llm_engine.py#L105-L111" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/engine/async_llm.py", "first": 149, "last": 156, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L149-L156" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/engine/core.py", "first": 583, "last": 596, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L583-L596" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 207, "last": 219, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L207-L219" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 453, "last": 468, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L453-L468" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 894, "last": 898, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L894-L898" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 2789, "last": 2802, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L2789-L2802" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/sampling_params.py", "first": 266, "last": 282, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L266-L282" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/sampling_params.py", "first": 337, "last": 344, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L337-L344" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/server_args.py", "first": 784, "last": 796, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L784-L796" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/entrypoints/llm.py", "first": 418, "last": 480, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/llm.py#L418-L480" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/entrypoints/offline_utils.py", "first": 326, "last": 349, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/offline_utils.py#L326-L349" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/engine/llm_engine.py", "first": 298, "last": 336, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/llm_engine.py#L298-L336" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 894, "last": 898, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L894-L898" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/server_args.py", "first": 887, "last": 900, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L887-L900" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3572, "last": 3589, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3572-L3589" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 577, "last": 581, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L577-L581" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 648, "last": 695, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L648-L695" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 2797, "last": 2797, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L2797-L2797" }, { "page": "00-foundations/00-01-what-inference-is.html", "path": "vllm/v1/engine/llm_engine.py", "first": 105, "last": 111, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/llm_engine.py#L105-L111" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 597, "last": 608, "engine": "vllm", "section": "what-is-cacheable", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L597-L608" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/attention/backend.py", "first": 485, "last": 485, "engine": "vllm", "section": "what-is-cacheable", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L485-L485" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/attention/backend.py", "first": 465, "last": 480, "engine": "vllm", "section": "what-is-cacheable", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L465-L480" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/kv_cache_interface.py", "first": 243, "last": 252, "engine": "vllm", "section": "memory-cost", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L243-L252" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "python/sglang/srt/model_executor/forward_batch_info.py", "first": 100, "last": 109, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/forward_batch_info.py#L100-L109" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 566, "last": 575, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L566-L575" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "python/sglang/srt/model_executor/forward_batch_info.py", "first": 1057, "last": 1061, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/forward_batch_info.py#L1057-L1061" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "python/sglang/srt/model_executor/forward_batch_info.py", "first": 871, "last": 875, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/forward_batch_info.py#L871-L875" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 2049, "last": 2060, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L2049-L2060" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 2105, "last": 2113, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L2105-L2113" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/attention/backend.py", "first": 89, "last": 96, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L89-L96" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 2331, "last": 2341, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L2331-L2341" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/engine/core.py", "first": 583, "last": 610, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L583-L610" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 566, "last": 575, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L566-L575" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/config/vllm.py", "first": 648, "last": 700, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L700" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/worker/gpu/input_batch.py", "first": 367, "last": 385, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/input_batch.py#L367-L385" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 4287, "last": 4291, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4287-L4291" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1416, "last": 1424, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1416-L1424" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2041, "last": 2055, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2041-L2055" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2248, "last": 2261, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2248-L2261" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/model_executor/models/llama.py", "first": 221, "last": 231, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L221-L231" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/model_executor/models/llama.py", "first": 157, "last": 158, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L157-L158" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 597, "last": 608, "engine": "vllm", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L597-L608" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "scheduler.py", "first": 566, "last": 575, "engine": null, "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/scheduler.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 35, "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "section": "production", "engine": "vllm", "path": "vllm/v1/core/sched/scheduler.py", "first": 566, "last": 575 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L566-L575" }, { "page": "00-foundations/00-02-decode-loop-and-kv-cache.html", "path": "docs/usage/v1_guide.md", "first": 28, "last": 29, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/usage/v1_guide.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/usage/v1_guide.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/usage/v1_guide.md#L28-L29" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/layernorm_kernels.cu", "first": 253, "last": 257, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/layernorm_kernels.cu#L253-L257" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/layernorm_kernels.cu", "first": 179, "last": 206, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/layernorm_kernels.cu#L179-L206" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/custom_all_reduce.cuh", "first": 7, "last": 23, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_reduce.cuh#L7-L23" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/cuda_compat.h", "first": 36, "last": 38, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cuda_compat.h#L36-L38" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/cuda_compat.h", "first": 46, "last": 55, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/cuda_compat.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/cuda_compat.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cuda_compat.h#L46-L55" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/attention_utils.cuh", "first": 38, "last": 47, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/attention_utils.cuh#L38-L47" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/layernorm_kernels.cu", "first": 267, "last": 271, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/layernorm_kernels.cu#L267-L271" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "first": 6, "last": 43, "engine": "vllm", "section": "tensor-cores-and-launch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_mma.h#L6-L43" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "vllm/model_executor/kernels/linear/mxfp8/rocm_native.py", "first": 133, "last": 137, "engine": "vllm", "section": "tensor-cores-and-launch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mxfp8/rocm_native.py#L133-L137" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "vllm/v1/worker/gpu/sample/states.py", "first": 76, "last": 92, "engine": "vllm", "section": "tensor-cores-and-launch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/sample/states.py#L76-L92" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "vllm/v1/cudagraph_dispatcher.py", "first": 15, "last": 32, "engine": "vllm", "section": "tensor-cores-and-launch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/cudagraph_dispatcher.py#L15-L32" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "python/sglang/srt/model_executor/cuda_graph_config.py", "first": 88, "last": 107, "engine": "sglang", "section": "tensor-cores-and-launch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/cuda_graph_config.py#L88-L107" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 307, "last": 313, "engine": "vllm", "section": "in-production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L307-L313" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 36, "last": 49, "engine": "vllm", "section": "in-production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L36-L49" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 268, "last": 268, "engine": null, "section": "in-production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L268-L268" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "python/sglang/kernels/ops/attention/merge_state.py", "first": 8, "last": 85, "engine": "sglang", "section": "in-production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/merge_state.py#L8-L85" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "python/sglang/srt/layers/attention/merge_state.py", "first": 34, "last": 46, "engine": "sglang", "section": "in-production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/merge_state.py#L34-L46" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 337, "last": 362, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L337-L362" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 221, "last": 237, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L221-L237" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/cuda_compat.h", "first": 70, "last": 76, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/cuda_compat.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/cuda_compat.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cuda_compat.h#L70-L76" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/cuda_utils.h", "first": 29, "last": 31, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/cuda_utils.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/cuda_utils.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cuda_utils.h#L29-L31" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/cuda_utils.h", "first": 29, "last": 31, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cuda_utils.h#L29-L31" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/attention_utils.cuh", "first": 38, "last": 47, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/attention_utils.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/attention_utils.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/attention_utils.cuh#L38-L47" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/cuda_compat.h", "first": 7, "last": 38, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/cuda_compat.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/cuda_compat.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cuda_compat.h#L7-L38" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "vllm/engine/arg_utils.py", "first": 880, "last": 880, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L880-L880" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "vllm/config/model.py", "first": 241, "last": 241, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L241-L241" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/layernorm_kernels.cu", "first": 253, "last": 271, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/layernorm_kernels.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/layernorm_kernels.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/layernorm_kernels.cu#L253-L271" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 307, "last": 313, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L307-L313" }, { "page": "00-foundations/00-03-gpu-architecture.html", "path": "csrc/cuda_compat.h", "first": 7, "last": 38, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/cuda_compat.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/cuda_compat.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cuda_compat.h#L7-L38" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/engine/arg_utils.py", "first": 2580, "last": 2671, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2580-L2671" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "benchmarks/kernels/weight_shapes.py", "first": 52, "last": 57, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "benchmarks/kernels/weight_shapes.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "benchmarks/kernels/weight_shapes.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/kernels/weight_shapes.py#L52-L57" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 502, "last": 510, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L502-L510" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 566, "last": 576, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L566-L576" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 704, "last": 710, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L704-L710" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/config/scheduler.py", "first": 42, "last": 55, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L42-L55" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2032, "last": 2050, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2032-L2050" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1032, "last": 1042, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1032-L1042" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/srt/server_args.py", "first": 826, "last": 836, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L826-L836" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 511, "last": 511, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L511-L511" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 841, "last": 860, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L841-L860" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "benchmarks/kernels/benchmark_fp8_gemm.py", "first": 89, "last": 123, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/kernels/benchmark_fp8_gemm.py#L89-L123" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/bench_one_batch.py", "first": 1, "last": 22, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/bench_one_batch.py#L1-L22" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "benchmarks/benchmark_latency.py", "first": 1, "last": 17, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "benchmarks/benchmark_latency.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "benchmarks/benchmark_latency.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_latency.py#L1-L17" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/benchmark/one_batch.py", "first": 784, "last": 804, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L784-L804" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/benchmark/one_batch.py", "first": 817, "last": 850, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L817-L850" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/benchmark/one_batch.py", "first": 487, "last": 521, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L487-L521" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/benchmark/one_batch.py", "first": 526, "last": 537, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L526-L537" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/benchmark/one_batch.py", "first": 784, "last": 804, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L784-L804" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/benchmark/one_batch.py", "first": 196, "last": 200, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L196-L200" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/benchmarks/latency.py", "first": 34, "last": 54, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L34-L54" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "vllm/benchmarks/latency.py", "first": 125, "last": 135, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L125-L135" }, { "page": "00-foundations/00-04-roofline-and-arithmetic-intensity.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 841, "last": 860, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L841-L860" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/config/model.py", "first": 2238, "last": 2259, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L2238-L2259" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/scalar_type.py", "first": 327, "last": 336, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/scalar_type.py#L327-L336" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/scalar_type.py", "first": 71, "last": 105, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/scalar_type.py#L71-L105" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/utils/fp8_utils.py", "first": 793, "last": 812, "engine": "vllm", "section": "accumulation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/fp8_utils.py#L793-L812" }, { "page": "00-foundations/00-05-numerics.html", "path": "python/sglang/srt/layers/layernorm.py", "first": 784, "last": 793, "engine": "sglang", "section": "accumulation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/layernorm.py#L784-L793" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/v1/worker/gpu/sample/sampler.py", "first": 209, "last": 210, "engine": "vllm", "section": "accumulation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/sample/sampler.py#L209-L210" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/v1/sample/sampler.py", "first": 96, "last": 97, "engine": "vllm", "section": "accumulation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L96-L97" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/awq_triton.py", "first": 96, "last": 102, "engine": "vllm", "section": "integer-and-block", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/awq_triton.py#L96-L102" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/utils/quant_utils.py", "first": 122, "last": 137, "engine": "vllm", "section": "integer-and-block", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/quant_utils.py#L122-L137" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/utils/ocp_mx_utils.py", "first": 9, "last": 18, "engine": "vllm", "section": "integer-and-block", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/ocp_mx_utils.py#L9-L18" }, { "page": "00-foundations/00-05-numerics.html", "path": "python/sglang/srt/layers/quantization/mxfp4_tensor.py", "first": 22, "last": 26, "engine": "sglang", "section": "integer-and-block", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/mxfp4_tensor.py#L22-L26" }, { "page": "00-foundations/00-05-numerics.html", "path": "python/sglang/srt/layers/quantization/mxfp4_tensor.py", "first": 61, "last": 74, "engine": "sglang", "section": "integer-and-block", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/mxfp4_tensor.py#L61-L74" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 12, "last": 46, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L12-L46" }, { "page": "00-foundations/00-05-numerics.html", "path": "python/sglang/srt/layers/quantization/__init__.py", "first": 72, "last": 101, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/__init__.py#L72-L101" }, { "page": "00-foundations/00-05-numerics.html", "path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "first": 48, "last": 79, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_dtype.py#L48-L79" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/config/cache.py", "first": 19, "last": 37, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L19-L37" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/v1/kv_cache_interface.py", "first": 36, "last": 48, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L36-L48" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/input_quant_fp8.py", "first": 199, "last": 218, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/input_quant_fp8.py#L199-L218" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/input_quant_fp8.py", "first": 199, "last": 218, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/input_quant_fp8.py#L199-L218" }, { "page": "00-foundations/00-05-numerics.html", "path": "csrc/quantization/w8a8/fp8/common.cuh", "first": 57, "last": 77, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/quantization/w8a8/fp8/common.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/quantization/w8a8/fp8/common.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/quantization/w8a8/fp8/common.cuh#L57-L77" }, { "page": "00-foundations/00-05-numerics.html", "path": "csrc/quantization/w8a8/fp8/common.cuh", "first": 57, "last": 77, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/quantization/w8a8/fp8/common.cuh#L57-L77" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/utils/fp8_utils.py", "first": 793, "last": 812, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/fp8_utils.py#L793-L812" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "first": 107, "last": 151, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L107-L151" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/config/model.py", "first": 2347, "last": 2351, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L2347-L2351" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/scalar_type.py", "first": 71, "last": 105, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/scalar_type.py#L71-L105" }, { "page": "00-foundations/00-05-numerics.html", "path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "first": 83, "last": 99, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_dtype.py#L83-L99" }, { "page": "00-foundations/00-05-numerics.html", "path": "vllm/scalar_type.py", "first": 338, "last": 339, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/scalar_type.py#L338-L339" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/model_executor/forward_batch_info.py", "first": 100, "last": 124, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/forward_batch_info.py#L100-L124" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/model_executor/forward_batch_info.py", "first": 150, "last": 183, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/forward_batch_info.py#L150-L183" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3188, "last": 3197, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3188-L3197" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 484, "last": 495, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L484-L495" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 565, "last": 574, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L565-L574" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/core/sched/output.py", "first": 205, "last": 222, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/output.py#L205-L222" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/worker/utils.py", "first": 601, "last": 607, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/utils.py#L601-L607" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3989, "last": 4007, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3989-L4007" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2041, "last": 2049, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2041-L2049" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 2758, "last": 2784, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L2758-L2784" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3513, "last": 3526, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3513-L3526" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/config/scheduler.py", "first": 74, "last": 74, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L74-L74" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/config/scheduler.py", "first": 42, "last": 44, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L42-L44" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/engine/arg_utils.py", "first": 2580, "last": 2671, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2580-L2671" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/server_args.py", "first": 4894, "last": 4899, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4894-L4899" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/config/scheduler.py", "first": 249, "last": 261, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L249-L261" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/managers/scheduler_components/metrics_reporter.py", "first": 604, "last": 612, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/metrics_reporter.py#L604-L612" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/metrics/loggers.py", "first": 261, "last": 266, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L261-L266" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 565, "last": 574, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L565-L574" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3513, "last": 3520, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3513-L3520" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 531, "last": 531, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L531-L531" }, { "page": "01-the-core-loop/01-01-prefill-vs-decode.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 484, "last": 495, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L484-L495" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 607, "last": 616, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L607-L616" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 726, "last": 734, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L726-L734" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 1268, "last": 1268, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1268-L1268" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "benchmarks/benchmark_serving.py", "first": 5, "last": 17, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_serving.py#L5-L17" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 197, "last": 257, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L197-L257" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 323, "last": 358, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L323-L358" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 607, "last": 611, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L607-L611" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 1134, "last": 1135, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1134-L1135" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 749, "last": 761, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L749-L761" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 1136, "last": 1147, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1136-L1147" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 1261, "last": 1261, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1261-L1261" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 1268, "last": 1268, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1268-L1268" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 622, "last": 645, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L622-L645" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 1432, "last": 1445, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1432-L1445" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/loggers.py", "first": 829, "last": 854, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L829-L854" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 1671, "last": 1683, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L1671-L1683" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/loggers.py", "first": 796, "last": 822, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L796-L822" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 1623, "last": 1642, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L1623-L1642" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/loggers.py", "first": 889, "last": 926, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L889-L926" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/loggers.py", "first": 859, "last": 886, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L859-L886" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 686, "last": 732, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L686-L732" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 733, "last": 739, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L733-L739" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 455, "last": 467, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L455-L467" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 1090, "last": 1090, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1090-L1090" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/engine/output_processor.py", "first": 183, "last": 183, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L183-L183" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/engine/output_processor.py", "first": 276, "last": 276, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L276-L276" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/stats.py", "first": 429, "last": 429, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L429-L429" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/stats.py", "first": 449, "last": 451, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L449-L451" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/stats.py", "first": 465, "last": 502, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L465-L502" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/stats.py", "first": 518, "last": 522, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L518-L522" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/stats.py", "first": 537, "last": 560, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L537-L560" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/loggers.py", "first": 1216, "last": 1219, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L1216-L1219" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/v1/metrics/loggers.py", "first": 1253, "last": 1255, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L1253-L1255" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 3369, "last": 3369, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L3369-L3369" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 2402, "last": 2403, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2402-L2403" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 2823, "last": 2841, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2823-L2841" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/observability/req_time_stats.py", "first": 473, "last": 483, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/req_time_stats.py#L473-L483" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 1822, "last": 1835, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L1822-L1835" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 964, "last": 972, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L964-L972" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/benchmark/serving.py", "first": 1380, "last": 1385, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1380-L1385" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 661, "last": 666, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L661-L666" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 595, "last": 604, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L595-L604" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 1807, "last": 1820, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L1807-L1820" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 964, "last": 972, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L964-L972" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 1435, "last": 1435, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1435-L1435" }, { "page": "01-the-core-loop/01-02-latency-metrics-and-slos.html", "path": "vllm/benchmarks/serve.py", "first": 1786, "last": 1788, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1786-L1788" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 250, "last": 266, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L250-L266" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 484, "last": 495, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L484-L495" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 531, "last": 534, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L531-L534" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 755, "last": 759, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L755-L759" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2060, "last": 2062, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2060-L2062" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2422, "last": 2442, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2422-L2442" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 324, "last": 336, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L324-L336" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 530, "last": 548, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L530-L548" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 708, "last": 786, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L708-L786" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/config/vllm.py", "first": 648, "last": 700, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L700" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu/states.py", "first": 9, "last": 133, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/states.py#L9-L133" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu/states.py", "first": 29, "last": 29, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/states.py#L29-L29" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu/states.py", "first": 126, "last": 133, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/states.py#L126-L133" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1083, "last": 1084, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1083-L1084" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1129, "last": 1129, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1129-L1129" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 2012, "last": 2024, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L2012-L2024" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1748, "last": 1779, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1748-L1779" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3142, "last": 3154, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3142-L3154" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3188, "last": 3197, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3188-L3197" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3563, "last": 3569, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3563-L3569" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 3147, "last": 3196, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L3147-L3196" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 3233, "last": 3252, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L3233-L3252" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 1302, "last": 1308, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L1302-L1308" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/engine/core.py", "first": 1391, "last": 1399, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1391-L1399" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/engine/core.py", "first": 583, "last": 611, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L583-L611" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2060, "last": 2062, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2060-L2062" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 1264, "last": 1271, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L1264-L1271" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 1562, "last": 1574, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L1562-L1574" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/scheduler_components/batch_result_processor.py", "first": 273, "last": 278, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/batch_result_processor.py#L273-L278" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 636, "last": 651, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L636-L651" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3618, "last": 3632, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3618-L3632" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 1286, "last": 1308, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L1286-L1308" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "vllm/config/scheduler.py", "first": 42, "last": 44, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L42-L44" }, { "page": "01-the-core-loop/01-03-continuous-batching.html", "path": "python/sglang/srt/server_args.py", "first": 789, "last": 791, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L789-L791" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3619, "last": 3631, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3619-L3631" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/metrics/loggers.py", "first": 283, "last": 285, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L283-L285" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/request.py", "first": 364, "last": 387, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L364-L387" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 759, "last": 766, "engine": "vllm", "section": "what-admission-tests", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L759-L766" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 524, "last": 530, "engine": "vllm", "section": "what-admission-tests", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L524-L530" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 466, "last": 473, "engine": "vllm", "section": "what-admission-tests", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L466-L473" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/config/scheduler.py", "first": 136, "last": 141, "engine": "vllm", "section": "what-admission-tests", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L136-L141" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/request_queue.py", "first": 13, "last": 17, "engine": "vllm", "section": "queues-and-policy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/request_queue.py#L13-L17" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/request.py", "first": 350, "last": 361, "engine": "vllm", "section": "queues-and-policy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L350-L361" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 200, "last": 213, "engine": "sglang", "section": "queues-and-policy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L200-L213" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 290, "last": 294, "engine": "sglang", "section": "queues-and-policy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L290-L294" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 505, "last": 508, "engine": "sglang", "section": "queues-and-policy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L505-L508" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 841, "last": 862, "engine": "sglang", "section": "queues-and-policy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L841-L862" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 661, "last": 668, "engine": "sglang", "section": "queues-and-policy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L661-L668" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 636, "last": 650, "engine": "vllm", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L636-L650" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 651, "last": 696, "engine": "vllm", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L651-L696" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1360, "last": 1388, "engine": "vllm", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1360-L1388" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 2825, "last": 2846, "engine": "sglang", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L2825-L2846" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 2892, "last": 2902, "engine": "sglang", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L2892-L2902" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 1437, "last": 1441, "engine": "sglang", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1437-L1441" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 756, "last": 756, "engine": "vllm", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L756-L756" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/block_pool.py", "first": 727, "last": 743, "engine": "vllm", "section": "recompute-vs-swap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L727-L743" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1924, "last": 1945, "engine": "sglang", "section": "recompute-vs-swap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1924-L1945" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/scheduler_components/new_token_ratio_tracker.py", "first": 34, "last": 51, "engine": "sglang", "section": "starvation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/new_token_ratio_tracker.py#L34-L51" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 630, "last": 632, "engine": "vllm", "section": "starvation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L630-L632" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 2878, "last": 2889, "engine": "sglang", "section": "starvation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L2878-L2889" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/metrics/stats.py", "first": 122, "last": 135, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L122-L135" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/config/scheduler.py", "first": 117, "last": 120, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L117-L120" }, { "page": "01-the-core-loop/01-04-the-scheduler.html", "path": "vllm/v1/core/sched/interface.py", "first": 54, "last": 62, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/interface.py#L54-L62" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/config/scheduler.py", "first": 227, "last": 236, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L227-L236" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1089, "last": 1089, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1089-L1089" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2302, "last": 2311, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2302-L2311" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2160, "last": 2165, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2160-L2165" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 484, "last": 497, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L484-L497" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 503, "last": 507, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L503-L507" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 935, "last": 940, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L935-L940" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 959, "last": 975, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L959-L975" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 566, "last": 575, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L566-L575" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1400, "last": 1412, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1400-L1412" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/config/scheduler.py", "first": 74, "last": 76, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L74-L76" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/engine/arg_utils.py", "first": 2678, "last": 2683, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2678-L2683" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/config/scheduler.py", "first": 41, "last": 43, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L41-L43" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/engine/arg_utils.py", "first": 2606, "last": 2628, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2606-L2628" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/engine/arg_utils.py", "first": 2629, "last": 2639, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2629-L2639" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/config/scheduler.py", "first": 70, "last": 72, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L70-L72" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3188, "last": 3196, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3188-L3196" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 537, "last": 546, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L537-L546" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 1386, "last": 1424, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1386-L1424" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3461, "last": 3467, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3461-L3467" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3341, "last": 3343, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3341-L3343" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 1036, "last": 1053, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1036-L1053" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/server_args.py", "first": 811, "last": 815, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L811-L815" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/server_args.py", "first": 4866, "last": 4925, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4866-L4925" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/server_args.py", "first": 996, "last": 1000, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L996-L1000" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 2758, "last": 2788, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L2758-L2788" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/config/scheduler.py", "first": 248, "last": 261, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L248-L261" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/config/scheduler.py", "first": 263, "last": 268, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L263-L268" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/server_args.py", "first": 9288, "last": 9291, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9288-L9291" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1170, "last": 1180, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1170-L1180" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/config/scheduler.py", "first": 241, "last": 246, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L241-L246" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/engine/arg_utils.py", "first": 2851, "last": 2856, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2851-L2856" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3466, "last": 3467, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3466-L3467" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/engine/arg_utils.py", "first": 2606, "last": 2639, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2606-L2639" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3463, "last": 3463, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3463-L3463" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "docs/configuration/optimization.md", "first": 80, "last": 80, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/configuration/optimization.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/configuration/optimization.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/configuration/optimization.md#L80-L80" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "docs/configuration/optimization.md", "first": 49, "last": 81, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/configuration/optimization.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/configuration/optimization.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/configuration/optimization.md#L49-L81" }, { "page": "01-the-core-loop/01-05-chunked-prefill.html", "path": "vllm/engine/arg_utils.py", "first": 2606, "last": 2608, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2606-L2608" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/server_args.py", "first": 3155, "last": 3159, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3155-L3159" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/utils.py", "first": 100, "last": 103, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/utils.py#L100-L103" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 569, "last": 604, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L569-L604" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/decode.py", "first": 2428, "last": 2463, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/decode.py#L2428-L2463" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/base/conn.py", "first": 93, "last": 98, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/base/conn.py#L93-L98" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/base/conn.py", "first": 207, "last": 217, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/base/conn.py#L207-L217" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/mooncake/conn.py", "first": 1971, "last": 1975, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/mooncake/conn.py#L1971-L1975" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/base.py", "first": 8, "last": 38, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/base.py#L8-L38" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/pull_scheduler.py", "first": 60, "last": 66, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/pull_scheduler.py#L60-L66" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/v1/request.py", "first": 364, "last": 371, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L364-L371" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2836, "last": 2864, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2836-L2864" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "first": 148, "last": 242, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/factory.py#L148-L242" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/model_executor/layers/attention/kv_transfer_utils.py", "first": 37, "last": 59, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/kv_transfer_utils.py#L37-L59" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/connector.py", "first": 326, "last": 334, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/connector.py#L326-L334" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_connector.py", "first": 580, "last": 588, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_connector.py#L580-L588" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/moriio/moriio_connector.py", "first": 2035, "last": 2058, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/moriio/moriio_connector.py#L2035-L2058" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 1059, "last": 1066, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L1059-L1066" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/mooncake/conn.py", "first": 648, "last": 655, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/mooncake/conn.py#L648-L655" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 1082, "last": 1120, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L1082-L1120" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/utils.py", "first": 589, "last": 594, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/utils.py#L589-L594" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/mooncake/utils.py", "first": 29, "last": 88, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/mooncake/utils.py#L29-L88" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/fake/conn.py", "first": 50, "last": 61, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/fake/conn.py#L50-L61" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/mori/conn.py", "first": 362, "last": 373, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/mori/conn.py#L362-L373" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "first": 176, "last": 222, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/factory.py#L176-L222" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "first": 135, "last": 141, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L135-L141" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 299, "last": 324, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L299-L324" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/decode.py", "first": 662, "last": 681, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/decode.py#L662-L681" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/decode.py", "first": 1490, "last": 1495, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/decode.py#L1490-L1495" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 343, "last": 356, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L343-L356" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/mooncake/conn.py", "first": 1617, "last": 1640, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/mooncake/conn.py#L1617-L1640" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/server_args.py", "first": 3198, "last": 3202, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3198-L3202" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 1310, "last": 1343, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L1310-L1343" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/common/conn.py", "first": 1282, "last": 1299, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/common/conn.py#L1282-L1299" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 937, "last": 964, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L937-L964" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/decode.py", "first": 2283, "last": 2295, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/decode.py#L2283-L2295" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/mooncake/conn.py", "first": 1996, "last": 2020, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/mooncake/conn.py#L1996-L2020" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/mooncake/conn.py", "first": 1003, "last": 1122, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/mooncake/conn.py#L1003-L1122" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/tp_mapping.py", "first": 38, "last": 56, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/tp_mapping.py#L38-L56" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 908, "last": 928, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L908-L928" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py", "first": 3, "last": 15, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py#L3-L15" }, { "page": "01-the-core-loop/01-06-pd-disaggregation.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 438, "last": 466, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L438-L466" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 798, "last": 809, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L798-L809" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/kv_cache_interface.py", "first": 381, "last": 392, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L381-L392" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 438, "last": 438, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L438-L438" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/config/cache.py", "first": 80, "last": 80, "engine": "vllm", "section": "how-many-fit", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L80-L80" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 352, "last": 352, "engine": "sglang", "section": "how-many-fit", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L352-L352" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/kv_cache_interface.py", "first": 243, "last": 259, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L243-L259" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/config/cache.py", "first": 59, "last": 59, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L59-L59" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/kv_cache_interface.py", "first": 300, "last": 305, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L300-L305" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1038, "last": 1040, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1038-L1040" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 326, "last": 333, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L326-L333" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/configs/model_config.py", "first": 1166, "last": 1177, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L1166-L1177" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "pool_configurator.py", "first": 257, "last": 265, "engine": null, "section": "production", "declared_engine": "sglang", "declared_path": "python/sglang/srt/model_executor/pool_configurator.py", "reviewed_symbol": "cell_size", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/model_executor/pool_configurator.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/model_executor/pool_configurator.py", "symbol": "cell_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L257-L265" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/mem_cache/kv_cache_configurator.py", "first": 2147, "last": 2153, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_configurator.py#L2147-L2153" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 415, "last": 424, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L415-L424" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 2056, "last": 2060, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L2056-L2060" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/server_args.py", "first": 5005, "last": 5020, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5005-L5020" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "server_args.py", "first": 4894, "last": 4903, "engine": null, "section": "production", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "chunked_prefill_size", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "chunked_prefill_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4894-L4903" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "server_args.py", "first": 5124, "last": 5124, "engine": null, "section": "production", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "reserved_mem", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "reserved_mem", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5124-L5124" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/worker/utils.py", "first": 444, "last": 451, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/utils.py#L444-L451" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 475, "last": 475, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L475-L475" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 559, "last": 562, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L559-L562" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 648, "last": 654, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L648-L654" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 967, "last": 989, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L967-L989" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1925, "last": 1931, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1925-L1931" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/mem_cache/kv_cache_configurator.py", "first": 1764, "last": 1811, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_configurator.py#L1764-L1811" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "memory_pool.py", "first": 1674, "last": 1701, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/memory_pool.py", "reviewed_symbol": "_finalize_allocation_log", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/memory_pool.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/memory_pool.py", "symbol": "_finalize_allocation_log", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L1674-L1701" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1094, "last": 1099, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1094-L1099" }, { "page": "02-memory-and-kv-cache/02-01-kv-cache-sizing.html", "path": "kv_cache_utils.py", "first": 821, "last": 840, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/core/kv_cache_utils.py", "reviewed_symbol": "estimate_max_model_len", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/kv_cache_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/kv_cache_utils.py", "symbol": "estimate_max_model_len", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L821-L840" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/block_pool.py", "first": 45, "last": 52, "engine": "vllm", "section": "where-the-os-analogy-breaks", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L45-L52" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 159, "last": 180, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L159-L180" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/block_pool.py", "first": 166, "last": 192, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L166-L192" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/block_pool.py", "first": 647, "last": 668, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L647-L668" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/block_pool.py", "first": 719, "last": 744, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L719-L744" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 91, "last": 94, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L91-L94" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/block_table.py", "first": 113, "last": 121, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L113-L121" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/block_table.py", "first": 157, "last": 173, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L157-L173" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/block_table.py", "first": 463, "last": 475, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L463-L475" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/utils.py", "first": 585, "last": 586, "engine": "vllm", "section": "the-three-data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/utils.py#L585-L586" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 393, "last": 404, "engine": "vllm", "section": "allocate-path", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L393-L404" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 524, "last": 530, "engine": "vllm", "section": "allocate-path", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L524-L530" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 637, "last": 643, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L637-L643" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 191, "last": 197, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L191-L197" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/sched/output.py", "first": 131, "last": 142, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/output.py#L131-L142" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 1526, "last": 1528, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L1526-L1528" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 96, "last": 99, "engine": "vllm", "section": "block-size", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L96-L99" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/config/cache.py", "first": 59, "last": 59, "engine": "vllm", "section": "block-size", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L59-L59" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "overrides.py", "first": 2395, "last": 2396, "engine": null, "section": "block-size", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/arg_groups/overrides.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 392, "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "section": "block-size", "engine": "sglang", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2389, "last": 2397 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2395-L2396" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/config/cache.py", "first": 56, "last": 63, "engine": "vllm", "section": "block-size", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L56-L63" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2389, "last": 2397, "engine": "sglang", "section": "block-size", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2389-L2397" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/block_table.py", "first": 29, "last": 49, "engine": "vllm", "section": "block-size", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L29-L49" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 129, "last": 139, "engine": "vllm", "section": "copy-on-write", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L129-L139" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 280, "last": 286, "engine": "vllm", "section": "copy-on-write", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L280-L286" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 344, "last": 354, "engine": "vllm", "section": "copy-on-write", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L344-L354" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 409, "last": 422, "engine": "vllm", "section": "copy-on-write", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L409-L422" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1048, "last": 1055, "engine": "vllm", "section": "copy-on-write", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1048-L1055" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/worker/utils.py", "first": 576, "last": 589, "engine": "vllm", "section": "copy-on-write", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/utils.py#L576-L589" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/block_pool.py", "first": 702, "last": 717, "engine": "vllm", "section": "copy-on-write", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L702-L717" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/attention/backend.py", "first": 465, "last": 483, "engine": "vllm", "section": "reaching-the-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L465-L483" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 259, "last": 265, "engine": "vllm", "section": "reaching-the-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L259-L265" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "python/sglang/srt/mem_cache/allocator/token.py", "first": 42, "last": 64, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/token.py#L42-L64" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 272, "last": 283, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L272-L283" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "python/sglang/srt/mem_cache/kv_cache_configurator.py", "first": 1702, "last": 1721, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_configurator.py#L1702-L1721" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "python/sglang/srt/mem_cache/allocator/paged.py", "first": 197, "last": 204, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/paged.py#L197-L204" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 799, "last": 806, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L799-L806" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/core/block_pool.py", "first": 659, "last": 659, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L659-L659" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "python/sglang/srt/mem_cache/allocation.py", "first": 160, "last": 169, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L160-L169" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "vllm/v1/metrics/loggers.py", "first": 195, "last": 195, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L195-L195" }, { "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "path": "single_type_kv_cache_manager.py", "first": 191, "last": 197, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/single_type_kv_cache_manager.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 385, "page": "02-memory-and-kv-cache/02-02-pagedattention.html", "section": "worked-trace", "engine": "vllm", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 191, "last": 197 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L191-L197" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 618, "last": 645, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L618-L645" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 88, "last": 106, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L88-L106" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 596, "last": 615, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L596-L615" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/config/cache.py", "first": 107, "last": 117, "engine": "vllm", "section": "in-vllm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L107-L117" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 745, "last": 767, "engine": "vllm", "section": "in-vllm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L745-L767" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/block_pool.py", "first": 198, "last": 223, "engine": "vllm", "section": "in-vllm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L198-L223" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 730, "last": 741, "engine": "vllm", "section": "in-vllm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L730-L741" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/block_pool.py", "first": 719, "last": 743, "engine": "vllm", "section": "in-vllm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L719-L743" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 235, "last": 242, "engine": "vllm", "section": "in-vllm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L235-L242" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 253, "last": 267, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L253-L267" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/sampling_params.py", "first": 539, "last": 543, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L539-L543" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/block_pool.py", "first": 773, "last": 788, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L773-L788" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 467, "last": 478, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L467-L478" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/metrics/loggers.py", "first": 584, "last": 602, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L584-L602" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/metrics/stats.py", "first": 122, "last": 142, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L122-L142" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "vllm/v1/core/kv_cache_coordinator.py", "first": 516, "last": 518, "engine": "vllm", "section": "the-tree-hook", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_coordinator.py#L516-L518" }, { "page": "02-memory-and-kv-cache/02-03-prefix-caching-hash.html", "path": "benchmarks/benchmark_prefix_caching.py", "first": 11, "last": 16, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "benchmarks/benchmark_prefix_caching.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "benchmarks/benchmark_prefix_caching.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_prefix_caching.py#L11-L16" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 143, "last": 143, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L143-L143" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 303, "last": 303, "engine": null, "section": "the-problem", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "class RadixCache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "class RadixCache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L303-L303" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "unified_radix_cache.py", "first": 148, "last": 148, "engine": null, "section": "the-problem", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "class UnifiedRadixCache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "class UnifiedRadixCache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L148-L148" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 238, "last": 251, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L238-L251" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/evict_policy.py", "first": 16, "last": 64, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/evict_policy.py#L16-L64" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 217, "last": 229, "engine": null, "section": "first-principles", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "child_key", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "child_key", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L217-L229" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 417, "last": 436, "engine": null, "section": "first-principles", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "page_aligned", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "page_aligned", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L417-L436" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 679, "last": 703, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L679-L703" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 181, "last": 215, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L181-L215" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 705, "last": 728, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L705-L728" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 758, "last": 791, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L758-L791" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 491, "last": 509, "engine": null, "section": "first-principles", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "prefix_len", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "prefix_len", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L491-L509" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 599, "last": 603, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 448, "page": "02-memory-and-kv-cache/02-04-radixattention.html", "section": "eviction", "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 593, "last": 621 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L599-L603" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "vllm/v1/core/block_pool.py", "first": 727, "last": 743, "engine": "vllm", "section": "tree-vs-hash", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L727-L743" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 733, "last": 740, "engine": "vllm", "section": "tree-vs-hash", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L733-L740" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 821, "last": 834, "engine": "sglang", "section": "tree-vs-hash", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L821-L834" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2378, "last": 2397, "engine": "sglang", "section": "granularity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2378-L2397" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "vllm/config/cache.py", "first": 59, "last": 59, "engine": "vllm", "section": "granularity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L59-L59" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 150, "last": 154, "engine": null, "section": "granularity", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "page_aligned", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "page_aligned", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L150-L154" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 593, "last": 621, "engine": "sglang", "section": "eviction", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L593-L621" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/evict_policy.py", "first": 16, "last": 46, "engine": "sglang", "section": "eviction", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/evict_policy.py#L16-L46" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/server_args.py", "first": 934, "last": 947, "engine": "sglang", "section": "eviction", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L934-L947" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 623, "last": 636, "engine": "sglang", "section": "eviction", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L623-L636" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 360, "last": 360, "engine": null, "section": "eviction", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "lock_ref = 1", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "lock_ref = 1", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L360-L360" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 200, "last": 214, "engine": "sglang", "section": "scheduler-coupling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L200-L214" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "vllm/v1/core/sched/request_queue.py", "first": 13, "last": 17, "engine": "vllm", "section": "scheduler-coupling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/request_queue.py#L13-L17" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 380, "last": 391, "engine": "sglang", "section": "scheduler-coupling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L380-L391" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 339, "last": 365, "engine": "sglang", "section": "scheduler-coupling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L339-L365" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 235, "last": 235, "engine": "sglang", "section": "scheduler-coupling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L235-L235" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 290, "last": 294, "engine": "sglang", "section": "scheduler-coupling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L290-L294" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/server_args.py", "first": 843, "last": 858, "engine": "sglang", "section": "scheduler-coupling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L843-L858" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1941, "last": 1945, "engine": "sglang", "section": "preemption", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1941-L1945" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/README.md", "first": 40, "last": 45, "engine": "sglang", "section": "variants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L40-L45" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 80, "last": 143, "engine": "sglang", "section": "variants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L80-L143" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/chunk_cache.py", "first": 67, "last": 77, "engine": "sglang", "section": "variants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/chunk_cache.py#L67-L77" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "python/sglang/srt/mem_cache/base_prefix_cache.py", "first": 273, "last": 336, "engine": "sglang", "section": "variants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/base_prefix_cache.py#L273-L336" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "schedule_policy.py", "first": 237, "last": 237, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/schedule_policy.py", "reviewed_symbol": "calc_priority", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "symbol": "calc_priority", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L237-L237" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "schedule_policy.py", "first": 138, "last": 197, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/schedule_policy.py", "reviewed_symbol": "match_prefix_for_req", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "symbol": "match_prefix_for_req", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L138-L197" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "schedule_policy.py", "first": 957, "last": 963, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/schedule_policy.py", "reviewed_symbol": "_req_inc_lock_ref", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "symbol": "_req_inc_lock_ref", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L957-L963" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 555, "last": 557, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "new_indices", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "new_indices", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L555-L557" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 586, "last": 588, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "pretty_print", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "pretty_print", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L586-L588" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "radix_cache.py", "first": 849, "last": 863, "engine": null, "section": "hands-on", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/radix_cache.py", "reviewed_symbol": "create_simulated", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/radix_cache.py", "symbol": "create_simulated", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L849-L863" }, { "page": "02-memory-and-kv-cache/02-04-radixattention.html", "path": "schedule_batch.py", "first": 1941, "last": 1941, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_batch.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 460, "page": "02-memory-and-kv-cache/02-04-radixattention.html", "section": "preemption", "engine": "sglang", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1941, "last": 1945 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1941-L1941" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/config/cache.py", "first": 80, "last": 80, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L80-L80" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "first": 57, "last": 69, "engine": "vllm", "section": "granularity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L57-L69" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache.py", "first": 124, "last": 127, "engine": null, "section": "granularity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/kv_cache.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 521, "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "section": "worked-trace", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "first": 74, "last": 152 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L124-L127" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "compressed_tensors.py", "first": 1129, "last": 1167, "engine": null, "section": "granularity", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "reviewed_symbol": "_tp_aware_loader", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "symbol": "_tp_aware_loader", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py#L1129-L1167" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/backends/triton_attn.py", "first": 277, "last": 289, "engine": "vllm", "section": "granularity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L277-L289" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/kv_cache_interface.py", "first": 244, "last": 260, "engine": "vllm", "section": "granularity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L244-L260" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "triton_attn.py", "first": 440, "last": 500, "engine": null, "section": "granularity", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/triton_attn.py", "reviewed_symbol": "_ensure_scale_caches", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/triton_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/triton_attn.py", "symbol": "_ensure_scale_caches", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L440-L500" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "fp4_kv_cache_quant_method.py", "first": 340, "last": 340, "engine": null, "section": "granularity", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py", "reviewed_symbol": "SCALE_BLOCK_SIZE", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py", "symbol": "SCALE_BLOCK_SIZE", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py#L340-L340" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/utils/torch_utils.py", "first": 531, "last": 534, "engine": "vllm", "section": "granularity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/torch_utils.py#L531-L534" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/config/cache.py", "first": 19, "last": 37, "engine": "vllm", "section": "dtypes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L19-L37" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "cache.py", "first": 95, "last": 97, "engine": null, "section": "dtypes", "declared_engine": "vllm", "declared_path": "vllm/config/cache.py", "reviewed_symbol": "nvfp4_4over6", "resolved_engine": "vllm", "resolved_path": "vllm/config/cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/config/cache.py", "symbol": "nvfp4_4over6", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L95-L97" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/utils/torch_utils.py", "first": 33, "last": 54, "engine": "vllm", "section": "dtypes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/torch_utils.py#L33-L54" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/kv_cache_interface.py", "first": 43, "last": 53, "engine": "vllm", "section": "dtypes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L43-L53" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "first": 37, "last": 81, "engine": "sglang", "section": "dtypes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_dtype.py#L37-L81" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache_dtype.py", "first": 37, "last": 45, "engine": null, "section": "dtypes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 485, "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "section": "dtypes", "engine": "sglang", "path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "first": 37, "last": 81 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_dtype.py#L37-L45" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 1649, "last": 1653, "engine": "sglang", "section": "dtypes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L1649-L1653" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 38, "last": 56, "engine": "vllm", "section": "dequant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L38-L56" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 385, "last": 386, "engine": "vllm", "section": "dequant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L385-L386" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "flash_attn.py", "first": 1016, "last": 1023, "engine": null, "section": "dequant", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flash_attn.py", "reviewed_symbol": "descale_shape", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flash_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flash_attn.py", "symbol": "descale_shape", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1016-L1023" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1003, "last": 1006, "engine": "vllm", "section": "dequant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1003-L1006" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "flashattention_backend.py", "first": 1288, "last": 1293, "engine": null, "section": "dequant", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/flashattention_backend.py", "reviewed_symbol": "k_scale", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/flashattention_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "symbol": "k_scale", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L1288-L1293" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 537, "last": 542, "engine": "vllm", "section": "dequant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L537-L542" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py", "first": 777, "last": 791, "engine": "sglang", "section": "dequant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py#L777-L791" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "fp4_kv_cache_quant_method.py", "first": 15, "last": 34, "engine": null, "section": "dequant", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py", "reviewed_symbol": "quantization", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py", "symbol": "quantization", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/fp4_kv_cache_quant_method.py#L15-L34" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/backend.py", "first": 379, "last": 380, "engine": "vllm", "section": "dequant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L379-L380" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "flash_attn.py", "first": 81, "last": 87, "engine": null, "section": "dequant", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flash_attn.py", "reviewed_symbol": "supported_kv_cache_dtypes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flash_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flash_attn.py", "symbol": "supported_kv_cache_dtypes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L81-L87" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "flashinfer.py", "first": 411, "last": 420, "engine": null, "section": "dequant", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "supported_kv_cache_dtypes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "supported_kv_cache_dtypes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L411-L420" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "triton_attn.py", "first": 296, "last": 306, "engine": null, "section": "dequant", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/triton_attn.py", "reviewed_symbol": "supported_kv_cache_dtypes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/triton_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/triton_attn.py", "symbol": "supported_kv_cache_dtypes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L296-L306" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "turboquant_attn.py", "first": 133, "last": 138, "engine": null, "section": "dequant", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/turboquant_attn.py", "reviewed_symbol": "supported_kv_cache_dtypes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/turboquant_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/turboquant_attn.py", "symbol": "supported_kv_cache_dtypes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/turboquant_attn.py#L133-L138" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/backends/fa_utils.py", "first": 223, "last": 243, "engine": "vllm", "section": "dequant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L223-L243" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "turboquant_attn.py", "first": 180, "last": 210, "engine": null, "section": "dequant", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/turboquant_attn.py", "reviewed_symbol": "get_kv_cache_shape", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/turboquant_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/turboquant_attn.py", "symbol": "get_kv_cache_shape", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/turboquant_attn.py#L180-L210" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/backends/turboquant_attn.py", "first": 105, "last": 120, "engine": "vllm", "section": "rotation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/turboquant_attn.py#L105-L120" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/model_executor/layers/quantization/turboquant/centroids.py", "first": 3, "last": 10, "engine": "vllm", "section": "rotation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/centroids.py#L3-L10" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/int4_per_token_head.py", "first": 1123, "last": 1135, "engine": "vllm", "section": "rotation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/int4_per_token_head.py#L1123-L1135" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/backends/turboquant_attn.py", "first": 438, "last": 445, "engine": "vllm", "section": "rotation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/turboquant_attn.py#L438-L445" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "turboquant_attn.py", "first": 489, "last": 491, "engine": null, "section": "rotation", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/turboquant_attn.py", "reviewed_symbol": "_build_hadamard", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/turboquant_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/turboquant_attn.py", "symbol": "_build_hadamard", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/turboquant_attn.py#L489-L491" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/model_executor/layers/quantization/turboquant/centroids.py", "first": 59, "last": 86, "engine": "vllm", "section": "rotation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/centroids.py#L59-L86" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/triton_turboquant_store.py", "first": 278, "last": 292, "engine": "vllm", "section": "rotation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_turboquant_store.py#L278-L292" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/model_executor/layers/quantization/turboquant/config.py", "first": 17, "last": 41, "engine": "vllm", "section": "rotated-schemes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/config.py#L17-L41" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "config.py", "first": 149, "last": 156, "engine": null, "section": "rotated-schemes", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "reviewed_symbol": "value_packed_size", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/turboquant/config.py", "symbol": "value_packed_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/config.py#L149-L156" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "config.py", "first": 94, "last": 97, "engine": null, "section": "rotated-schemes", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "reviewed_symbol": "key_fp8", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/turboquant/config.py", "symbol": "key_fp8", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/config.py#L94-L97" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "config.py", "first": 83, "last": 85, "engine": null, "section": "rotated-schemes", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "reviewed_symbol": "norm_correction", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/turboquant/config.py", "symbol": "norm_correction", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/config.py#L83-L85" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/model_executor/layers/quantization/turboquant/config.py", "first": 188, "last": 192, "engine": "vllm", "section": "rotated-schemes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/config.py#L188-L192" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/config/cache.py", "first": 134, "last": 136, "engine": "vllm", "section": "rotated-schemes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L134-L136" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/int4_per_token_head.py", "first": 3, "last": 11, "engine": "vllm", "section": "rotated-schemes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/int4_per_token_head.py#L3-L11" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "int4_per_token_head.py", "first": 157, "last": 161, "engine": null, "section": "rotated-schemes", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/ops/int4_per_token_head.py", "reviewed_symbol": "k_zp_int", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/ops/int4_per_token_head.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/ops/int4_per_token_head.py", "symbol": "k_zp_int", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/int4_per_token_head.py#L157-L161" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/int4_per_token_head.py", "first": 567, "last": 574, "engine": "vllm", "section": "rotated-schemes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/int4_per_token_head.py#L567-L574" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/ops/int4_per_token_head.py", "first": 909, "last": 918, "engine": "vllm", "section": "rotated-schemes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/int4_per_token_head.py#L909-L918" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/config/cache.py", "first": 293, "last": 311, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L293-L311" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "first": 74, "last": 152, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L74-L152" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/kv_cache_interface.py", "first": 254, "last": 260, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L254-L260" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "flash_attn.py", "first": 1003, "last": 1006, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flash_attn.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 491, "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "section": "dequant", "engine": "vllm", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1003, "last": 1006 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1003-L1006" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "flash_attn.py", "first": 1016, "last": 1023, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flash_attn.py", "reviewed_symbol": "descale_shape", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flash_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flash_attn.py", "symbol": "descale_shape", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1016-L1023" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache.py", "first": 82, "last": 91, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/kv_cache.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 521, "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "section": "worked-trace", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "first": 74, "last": 152 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L82-L91" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "triton_reshape_and_cache_flash.py", "first": 207, "last": 214, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/ops/triton_reshape_and_cache_flash.py", "reviewed_symbol": "k_scale", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/ops/triton_reshape_and_cache_flash.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/ops/triton_reshape_and_cache_flash.py", "symbol": "k_scale", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_reshape_and_cache_flash.py#L207-L214" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "turboquant/__init__.py", "first": 3, "last": 22, "engine": null, "section": "accuracy", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/turboquant/__init__.py", "reviewed_symbol": "TurboQuant", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/turboquant/__init__.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/turboquant/__init__.py", "symbol": "TurboQuant", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/__init__.py#L3-L22" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "turboquant/config.py", "first": 71, "last": 75, "engine": null, "section": "accuracy", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "reviewed_symbol": "presets", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/turboquant/config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/turboquant/config.py", "symbol": "presets", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/turboquant/config.py#L71-L75" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache.py", "first": 145, "last": 151, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/kv_cache.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 521, "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "section": "worked-trace", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "first": 74, "last": 152 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L145-L151" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache.py", "first": 184, "last": 190, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/kv_cache.py", "reviewed_symbol": "kv_cache_dtype", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/kv_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "symbol": "kv_cache_dtype", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L184-L190" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache.py", "first": 130, "last": 134, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/kv_cache.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 521, "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "section": "worked-trace", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/kv_cache.py", "first": 74, "last": 152 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/kv_cache.py#L130-L134" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/v1/attention/backend.py", "first": 379, "last": 380, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L379-L380" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "vllm/config/cache.py", "first": 140, "last": 144, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L140-L144" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache_dtype.py", "first": 83, "last": 99, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "reviewed_symbol": "DFLASH", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/kv_cache_dtype.py", "symbol": "DFLASH", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_dtype.py#L83-L99" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "tests/evals/gsm8k/test_gsm8k_correctness.py", "first": 8, "last": 9, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/test_gsm8k_correctness.py#L8-L9" }, { "page": "02-memory-and-kv-cache/02-05-kv-quantization.html", "path": "kv_cache_interface.py", "first": 80, "last": 94, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/kv_cache_interface.py", "reviewed_symbol": "get_kv_quant_mode", "resolved_engine": "vllm", "resolved_path": "vllm/v1/kv_cache_interface.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/kv_cache_interface.py", "symbol": "get_kv_quant_mode", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L80-L94" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/config/cache.py", "first": 80, "last": 87, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L80-L87" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/utils.py", "first": 444, "last": 464, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/utils.py#L444-L464" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 513, "last": 519, "engine": "vllm", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L513-L519" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 6612, "last": 6615, "engine": "vllm", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L6612-L6615" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/utils/mem_utils.py", "first": 314, "last": 326, "engine": "vllm", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/mem_utils.py#L314-L326" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 6775, "last": 6797, "engine": "vllm", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L6775-L6797" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 559, "last": 563, "engine": "vllm", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L559-L563" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1023, "last": 1041, "engine": "vllm", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1023-L1041" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/server_args.py", "first": 4992, "last": 5012, "engine": "sglang", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4992-L5012" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/server_args.py", "first": 4839, "last": 4862, "engine": "sglang", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4839-L4862" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/kv_cache_configurator.py", "first": 1764, "last": 1775, "engine": "sglang", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_configurator.py#L1764-L1775" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/server_args.py", "first": 5043, "last": 5067, "engine": "sglang", "section": "how-they-profile", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5043-L5067" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 761, "last": 770, "engine": "vllm", "section": "why-098-oom", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L761-L770" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/config/cache.py", "first": 201, "last": 208, "engine": "vllm", "section": "why-098-oom", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L201-L208" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/utils.py", "first": 453, "last": 461, "engine": "vllm", "section": "debugging", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/utils.py#L453-L461" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 778, "last": 784, "engine": "vllm", "section": "debugging", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L778-L784" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 799, "last": 807, "engine": "vllm", "section": "debugging", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L799-L807" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 550, "last": 558, "engine": "vllm", "section": "debugging", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L550-L558" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/kv_cache_configurator.py", "first": 1800, "last": 1809, "engine": "sglang", "section": "debugging", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_configurator.py#L1800-L1809" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/pool_host/base.py", "first": 173, "last": 180, "engine": "sglang", "section": "debugging", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/pool_host/base.py#L173-L180" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 80, "last": 143, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L80-L143" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 187, "last": 196, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L187-L196" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 230, "last": 235, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L230-L235" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/README.md", "first": 38, "last": 45, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L38-L45" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/hybrid_cache/hybrid_cache_controller.py", "first": 96, "last": 96, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hybrid_cache/hybrid_cache_controller.py#L96-L96" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/pool_host/base.py", "first": 149, "last": 162, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/pool_host/base.py#L149-L162" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/server_args.py", "first": 7426, "last": 7440, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L7426-L7440" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/managers/cache_controller.py", "first": 334, "last": 339, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/cache_controller.py#L334-L339" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 443, "last": 451, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L443-L451" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 847, "last": 860, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L847-L860" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1244, "last": 1250, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1244-L1250" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/hiradix_cache.py", "first": 1231, "last": 1233, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hiradix_cache.py#L1231-L1233" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/hiradix_cache.py", "first": 841, "last": 848, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hiradix_cache.py#L841-L848" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1907, "last": 1923, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1907-L1923" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/hicache_storage.py", "first": 150, "last": 155, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hicache_storage.py#L150-L155" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/storage/backend_factory.py", "first": 196, "last": 251, "engine": "sglang", "section": "hicache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/storage/backend_factory.py#L196-L251" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 1587, "last": 1620, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L1587-L1620" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/hicache_storage.py", "first": 165, "last": 196, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hicache_storage.py#L165-L196" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 2185, "last": 2191, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L2185-L2191" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 2135, "last": 2154, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L2135-L2154" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 1754, "last": 1810, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L1754-L1810" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1786, "last": 1794, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1786-L1794" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 1396, "last": 1405, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L1396-L1405" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/l2_transfer.py", "first": 85, "last": 110, "engine": "sglang", "section": "prefetch-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/l2_transfer.py#L85-L110" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/kv_offload/base.py", "first": 163, "last": 184, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_offload/base.py#L163-L184" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/kv_offload/factory.py", "first": 61, "last": 69, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_offload/factory.py#L61-L69" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/kv_offload/cpu/policies/factory.py", "first": 84, "last": 90, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_offload/cpu/policies/factory.py#L84-L90" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/kv_offload/cpu/spec.py", "first": 90, "last": 110, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_offload/cpu/spec.py#L90-L110" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/kv_offload/tiering/spec.py", "first": 84, "last": 95, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_offload/tiering/spec.py#L84-L95" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/kv_offload/tiering/factory.py", "first": 103, "last": 125, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_offload/tiering/factory.py#L103-L125" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/simple_kv_offload/manager.py", "first": 94, "last": 142, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/simple_kv_offload/manager.py#L94-L142" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/simple_kv_offload/disk_backend.py", "first": 3, "last": 8, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/simple_kv_offload/disk_backend.py#L3-L8" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/v1/simple_kv_offload/manager.py", "first": 474, "last": 489, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/simple_kv_offload/manager.py#L474-L489" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "vllm/config/cache.py", "first": 210, "last": 219, "engine": "vllm", "section": "vllm-offload", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L210-L219" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1094, "last": 1099, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1094-L1099" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/pool_host/base.py", "first": 173, "last": 180, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/pool_host/base.py#L173-L180" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/hiradix_cache.py", "first": 841, "last": 848, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hiradix_cache.py#L841-L848" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1907, "last": 1923, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1907-L1923" }, { "page": "02-memory-and-kv-cache/02-06-offload-and-vram-budget.html", "path": "python/sglang/srt/mem_cache/README.md", "first": 38, "last": 45, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L38-L45" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 854, "last": 857, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L854-L857" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "first": 1535, "last": 1535, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L1535-L1535" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 15, "last": 15, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L15-L15" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 148, "last": 171, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L148-L171" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 106, "last": 106, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L106-L106" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 21, "last": 22, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L21-L22" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 195, "last": 200, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L195-L200" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1794, "last": 1794, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1794-L1794" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "flash_attn.py", "first": 1384, "last": 1384, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flash_attn.py", "reviewed_symbol": "merge_attn_states", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flash_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flash_attn.py", "symbol": "merge_attn_states", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1384-L1384" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/backends/flashinfer.py", "first": 384, "last": 384, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L384-L384" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 685, "last": 697, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L685-L697" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "python/sglang/srt/layers/attention/merge_state.py", "first": 34, "last": 46, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/merge_state.py#L34-L46" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "python/sglang/kernels/ops/attention/merge_state.py", "first": 25, "last": 37, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/merge_state.py#L25-L37" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "python/sglang/kernels/ops/attention/merge_state.py", "first": 56, "last": 63, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/merge_state.py#L56-L63" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/merge_attn_states.py", "first": 80, "last": 100, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/merge_attn_states.py#L80-L100" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "python/sglang/srt/layers/attention/merge_state.py", "first": 19, "last": 23, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/merge_state.py#L19-L23" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/merge_attn_states.py", "first": 70, "last": 78, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/merge_attn_states.py#L70-L78" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1771, "last": 1791, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1771-L1791" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1793, "last": 1794, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1793-L1794" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/merge_attn_states.py", "first": 80, "last": 100, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/merge_attn_states.py#L80-L100" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 44, "last": 49, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L44-L49" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 103, "last": 116, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L103-L116" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/triton_merge_attn_states.py", "first": 160, "last": 164, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_merge_attn_states.py#L160-L164" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "python/sglang/kernels/ops/attention/merge_state.py", "first": 27, "last": 28, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/merge_state.py#L27-L28" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/triton_merge_attn_states.py", "first": 176, "last": 178, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_merge_attn_states.py#L176-L178" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/triton_merge_attn_states.py", "first": 201, "last": 206, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_merge_attn_states.py#L201-L206" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1790, "last": 1790, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1790-L1790" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1790, "last": 1790, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1790-L1790" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "merge_attn_states.cu", "first": 196, "last": 196, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 602, "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "section": "production", "engine": "vllm", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 195, "last": 200 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L196-L196" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "triton_merge_attn_states.py", "first": 207, "last": 209, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/ops/triton_merge_attn_states.py", "reviewed_symbol": "max_lse", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/ops/triton_merge_attn_states.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/ops/triton_merge_attn_states.py", "symbol": "max_lse", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_merge_attn_states.py#L207-L209" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 15, "last": 15, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L15-L15" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "vllm/v1/attention/ops/merge_attn_states.py", "first": 22, "last": 22, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/merge_attn_states.py#L22-L22" }, { "page": "03-attention-kernels/03-01-naive-attention-and-online-softmax.html", "path": "csrc/cpu/mla_decode.cpp", "first": 129, "last": 129, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/cpu/mla_decode.cpp", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/cpu/mla_decode.cpp", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cpu/mla_decode.cpp#L129-L129" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 176, "last": 207, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L176-L207" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 240, "last": 268, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L240-L268" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 224, "last": 234, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L224-L234" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1123, "last": 1145, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1123-L1145" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 276, "last": 278, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L276-L278" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 256, "last": 265, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L256-L265" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/model_executor/models/llama.py", "first": 159, "last": 159, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L159-L159" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/fa_utils.py", "first": 91, "last": 100, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L91-L100" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 62, "last": 69, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L62-L69" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "first": 260, "last": 300, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L260-L300" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "python/sglang/srt/layers/attention/attention_registry.py", "first": 209, "last": 241, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L209-L241" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "python/sglang/srt/server_args.py", "first": 5925, "last": 5934, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5925-L5934" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "first": 1479, "last": 1494, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L1479-L1494" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "first": 1496, "last": 1514, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L1496-L1514" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 538, "last": 740, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L538-L740" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 447, "last": 447, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L447-L447" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 147, "last": 172, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L147-L172" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 141, "last": 151, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L141-L151" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "flash_attn.py", "first": 983, "last": 984, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flash_attn.py", "reviewed_symbol": "key_cache", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flash_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flash_attn.py", "symbol": "key_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L983-L984" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "flash_attn.py", "first": 1016, "last": 1016, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flash_attn.py", "reviewed_symbol": "descale_shape", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flash_attn.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flash_attn.py", "symbol": "descale_shape", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1016-L1016" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 349, "last": 385, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L349-L385" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 269, "last": 275, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L269-L275" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 868, "last": 871, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L868-L871" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 985, "last": 987, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L985-L987" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/fa_utils.py", "first": 223, "last": 243, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L223-L243" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "vllm/v1/attention/backends/fa_utils.py", "first": 102, "last": 110, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L102-L110" }, { "page": "03-attention-kernels/03-02-flashattention.html", "path": "fa_utils.py", "first": 19, "last": 24, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/fa_utils.py", "reviewed_symbol": "_custom_ops", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/fa_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/fa_utils.py", "symbol": "_custom_ops", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L19-L24" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 929, "last": 936, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L929-L936" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 1076, "last": 1082, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L1076-L1082" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 966, "last": 966, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L966-L966" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 322, "last": 327, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L322-L327" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 798, "last": 799, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L798-L799" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 338, "last": 341, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L338-L341" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 305, "last": 307, "engine": "vllm", "section": "split-k-structure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L305-L307" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/backends/triton_attn.py", "first": 156, "last": 177, "engine": "vllm", "section": "split-k-structure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L156-L177" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 735, "last": 760, "engine": "vllm", "section": "split-k-structure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L735-L760" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 421, "last": 426, "engine": "vllm", "section": "paged-gather", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L421-L426" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 474, "last": 479, "engine": "vllm", "section": "paged-gather", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L474-L479" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 375, "last": 375, "engine": "vllm", "section": "paged-gather", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L375-L375" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/backends/triton_attn.py", "first": 308, "last": 310, "engine": "vllm", "section": "paged-gather", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L308-L310" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/worker/block_table.py", "first": 239, "last": 266, "engine": "vllm", "section": "paged-gather", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L239-L266" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/worker/block_table.py", "first": 29, "last": 49, "engine": "vllm", "section": "paged-gather", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L29-L49" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/backends/triton_attn.py", "first": 54, "last": 56, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L54-L56" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 1036, "last": 1050, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L1036-L1050" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "python/sglang/kernels/ops/attention/metadata.py", "first": 36, "last": 55, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/metadata.py#L36-L55" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "python/sglang/srt/server_args.py", "first": 1988, "last": 1992, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1988-L1992" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "python/sglang/kernels/ops/attention/decode_attention.py", "first": 845, "last": 846, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/decode_attention.py#L845-L846" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "python/sglang/kernels/ops/attention/decode_attention.py", "first": 40, "last": 41, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/decode_attention.py#L40-L41" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "python/sglang/kernels/ops/attention/decode_attention.py", "first": 664, "last": 674, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/decode_attention.py#L664-L674" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "csrc/rocm/attention.cu", "first": 3281, "last": 3290, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/rocm/attention.cu#L3281-L3290" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "csrc/rocm/attention.cu", "first": 3295, "last": 3300, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/rocm/attention.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/rocm/attention.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/rocm/attention.cu#L3295-L3300" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 15, "last": 16, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L15-L16" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/chunked_prefill_paged_decode.py", "first": 392, "last": 430, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/chunked_prefill_paged_decode.py#L392-L430" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/backends/triton_attn.py", "first": 197, "last": 271, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L197-L271" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/backends/triton_attn.py", "first": 599, "last": 749, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/triton_attn.py#L599-L749" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 802, "last": 802, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L802-L802" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_attention_helpers.py", "first": 231, "last": 236, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_attention_helpers.py#L231-L236" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 587, "last": 642, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L587-L642" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 1169, "last": 1189, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L1169-L1189" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 34, "last": 34, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L34-L34" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 580, "last": 593, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L580-L593" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "first": 408, "last": 419, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L408-L419" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "python/sglang/srt/layers/attention/triton_backend.py", "first": 376, "last": 385, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/triton_backend.py#L376-L385" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "vllm/v1/worker/block_table.py", "first": 39, "last": 43, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L39-L43" }, { "page": "03-attention-kernels/03-03-decode-attention.html", "path": "benchmarks/kernels/benchmark_paged_attention.py", "first": 145, "last": 151, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/kernels/benchmark_paged_attention.py#L145-L151" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/platforms/cuda.py", "first": 414, "last": 433, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L414-L433" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 353, "last": 440, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L353-L440" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 55, "last": 92, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L55-L92" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 250, "last": 335, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L250-L335" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 672, "last": 688, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L672-L688" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 743, "last": 762, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L743-L762" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 465, "last": 483, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L465-L483" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2620, "last": 2628, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2620-L2628" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 680, "last": 688, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L680-L688" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 7160, "last": 7175, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L7160-L7175" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/layers/attention/base_attn_backend.py", "first": 36, "last": 57, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/base_attn_backend.py#L36-L57" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "first": 1414, "last": 1416, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L1414-L1416" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/registry.py", "first": 44, "last": 68, "engine": "vllm", "section": "the-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L44-L68" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "registry.py", "first": 130, "last": 130, "engine": null, "section": "the-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/registry.py", "reviewed_symbol": "CUSTOM", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/registry.py", "symbol": "CUSTOM", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L130-L130" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "registry.py", "first": 243, "last": 295, "engine": null, "section": "the-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/registry.py", "reviewed_symbol": "register_backend", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/registry.py", "symbol": "register_backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L243-L295" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "registry.py", "first": 176, "last": 193, "engine": null, "section": "the-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/registry.py", "reviewed_symbol": "MambaAttentionBackendEnum", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/registry.py", "symbol": "MambaAttentionBackendEnum", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L176-L193" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/fa_utils.py", "first": 91, "last": 100, "engine": "vllm", "section": "the-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L91-L100" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "fa_utils.py", "first": 100, "last": 175, "engine": null, "section": "the-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/fa_utils.py", "reviewed_symbol": "fa_version", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/fa_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/fa_utils.py", "symbol": "fa_version", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L100-L175" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "first": 22, "last": 45, "engine": "vllm", "section": "the-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L22-L45" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flash_attn_interface.py", "first": 388, "last": 402, "engine": null, "section": "the-inventory", "declared_engine": "vllm", "declared_path": "vllm/vllm_flash_attn/flash_attn_interface.py", "reviewed_symbol": "fa_version", "resolved_engine": "vllm", "resolved_path": "vllm/vllm_flash_attn/flash_attn_interface.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/vllm_flash_attn/flash_attn_interface.py", "symbol": "fa_version", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/vllm_flash_attn/flash_attn_interface.py#L388-L402" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "attention_registry.py", "first": 235, "last": 241, "engine": null, "section": "the-inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/attention_registry.py", "reviewed_symbol": "create_flashattention_v4_backend", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/attention_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/attention_registry.py", "symbol": "create_flashattention_v4_backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L235-L241" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "first": 260, "last": 300, "engine": "sglang", "section": "the-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L260-L300" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1123, "last": 1146, "engine": "vllm", "section": "the-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1123-L1146" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashattention_backend.py", "first": 1496, "last": 1514, "engine": null, "section": "the-inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/flashattention_backend.py", "reviewed_symbol": "result", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/flashattention_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "symbol": "result", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L1496-L1514" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/server_args.py", "first": 181, "last": 209, "engine": "sglang", "section": "the-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L181-L209" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/layers/attention/attention_registry.py", "first": 31, "last": 39, "engine": "sglang", "section": "the-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L31-L39" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "attention_registry.py", "first": 42, "last": 66, "engine": null, "section": "the-inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/attention_registry.py", "reviewed_symbol": "create_flashinfer_backend", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/attention_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/attention_registry.py", "symbol": "create_flashinfer_backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L42-L66" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/engine/arg_utils.py", "first": 975, "last": 977, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L975-L977" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/config/attention.py", "first": 24, "last": 25, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/attention.py#L24-L25" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/selector.py", "first": 105, "last": 192, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L105-L192" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/selector.py", "first": 176, "last": 192, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L176-L192" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/platforms/cuda.py", "first": 144, "last": 163, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L144-L163" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/platforms/cuda.py", "first": 455, "last": 459, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L455-L459" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/server_args.py", "first": 5925, "last": 5956, "engine": "sglang", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5925-L5956" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "attention_registry.py", "first": 209, "last": 232, "engine": null, "section": "selection", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/attention_registry.py", "reviewed_symbol": "create_flashattention_v3_backend", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/attention_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/attention_registry.py", "symbol": "create_flashattention_v3_backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L209-L232" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py", "first": 193, "last": 212, "engine": "sglang", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py#L193-L212" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/flashinfer.py", "first": 1255, "last": 1310, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L1255-L1310" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/flashinfer.py", "first": 1632, "last": 1651, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L1632-L1651" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "first": 286, "last": 286, "engine": "sglang", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L286-L286" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "first": 209, "last": 216, "engine": "sglang", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L209-L216" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer_backend.py", "first": 478, "last": 480, "engine": null, "section": "flashinfer", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "reviewed_symbol": "prefill_wrapper_ragged", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "symbol": "prefill_wrapper_ragged", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L478-L480" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer.py", "first": 1519, "last": 1541, "engine": null, "section": "flashinfer", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "cascade_wrapper", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "cascade_wrapper", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L1519-L1541" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer.py", "first": 282, "last": 392, "engine": null, "section": "flashinfer", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "BatchDCPPrefillWrapper", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "BatchDCPPrefillWrapper", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L282-L392" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/flashinfer.py", "first": 1742, "last": 1751, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L1742-L1751" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/utils/flashinfer.py", "first": 445, "last": 530, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/flashinfer.py#L445-L530" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/utils/flashinfer.py", "first": 511, "last": 525, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/flashinfer.py#L511-L525" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/utils/flashinfer.py", "first": 47, "last": 63, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/flashinfer.py#L47-L63" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/utils/flashinfer.py", "first": 367, "last": 391, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/flashinfer.py#L367-L391" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/model_executor/warmup/kernel_warmup.py", "first": 123, "last": 137, "engine": "vllm", "section": "flashinfer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/warmup/kernel_warmup.py#L123-L137" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 655, "last": 669, "engine": "vllm", "section": "cuda-graphs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L655-L669" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/flashinfer.py", "first": 1215, "last": 1232, "engine": "vllm", "section": "cuda-graphs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L1215-L1232" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer_backend.py", "first": 1046, "last": 1059, "engine": null, "section": "cuda-graphs", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "reviewed_symbol": "_create_decode_wrappers", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "symbol": "_create_decode_wrappers", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L1046-L1059" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "base_attn_backend.py", "first": 124, "last": 129, "engine": null, "section": "cuda-graphs", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/base_attn_backend.py", "reviewed_symbol": "attention", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/base_attn_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/base_attn_backend.py", "symbol": "attention", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/base_attn_backend.py#L124-L129" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 409, "last": 421, "engine": "vllm", "section": "cuda-graphs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L409-L421" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "selector.py", "first": 105, "last": 105, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/selector.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 723, "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "section": "selection", "engine": "vllm", "path": "vllm/v1/attention/selector.py", "first": 105, "last": 192 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L105-L105" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "gpu_model_runner.py", "first": 7150, "last": 7150, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "initialize_attn_backend", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "initialize_attn_backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L7150-L7150" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer.py", "first": 701, "last": 701, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "__init__", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "__init__", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L701-L701" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "platforms/cuda.py", "first": 470, "last": 488, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/platforms/cuda.py", "reviewed_symbol": "specified", "resolved_engine": "vllm", "resolved_path": "vllm/platforms/cuda.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/platforms/cuda.py", "symbol": "specified", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L470-L488" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer.py", "first": 1394, "last": 1398, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "has_same_window_lefts", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "has_same_window_lefts", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L1394-L1398" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer.py", "first": 899, "last": 908, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "NotImplementedError", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "NotImplementedError", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L899-L908" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "attention_registry.py", "first": 212, "last": 217, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/attention_registry.py", "reviewed_symbol": "get_device_capability", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/attention_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/attention_registry.py", "symbol": "get_device_capability", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L212-L217" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "attention_registry.py", "first": 392, "last": 401, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/attention_registry.py", "reviewed_symbol": "prefill_attention_backend_str", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/attention_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/attention_registry.py", "symbol": "prefill_attention_backend_str", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L392-L401" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "platforms/cuda.py", "first": 490, "last": 498, "engine": null, "section": "hands-on", "declared_engine": "vllm", "declared_path": "vllm/platforms/cuda.py", "reviewed_symbol": "logger", "resolved_engine": "vllm", "resolved_path": "vllm/platforms/cuda.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/platforms/cuda.py", "symbol": "logger", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L490-L498" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/v1/attention/backend.py", "first": 353, "last": 440, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L353-L440" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "flashinfer.py", "first": 949, "last": 985, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "get_q_data_type", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "get_q_data_type", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L949-L985" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "vllm/utils/flashinfer.py", "first": 511, "last": 525, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/flashinfer.py#L511-L525" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "attention_backend_setup.py", "first": 181, "last": 235, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py", "reviewed_symbol": "_build_resolved_backend", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py", "symbol": "_build_resolved_backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py#L181-L235" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "cuda.py", "first": 420, "last": 420, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/platforms/cuda.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 694, "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "section": "the-problem", "engine": "vllm", "path": "vllm/platforms/cuda.py", "first": 414, "last": 433 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L420-L420" }, { "page": "03-attention-kernels/03-04-flashinfer-and-backends.html", "path": "cuda.py", "first": 389, "last": 395, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/platforms/cuda.py", "reviewed_symbol": "invalid_reasons_i", "resolved_engine": "vllm", "resolved_path": "vllm/platforms/cuda.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/platforms/cuda.py", "symbol": "invalid_reasons_i", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L389-L395" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/configs/model_config.py", "first": 85, "last": 87, "engine": "sglang", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L85-L87" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/transformers_utils/model_arch_config_convertor.py", "first": 147, "last": 164, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/transformers_utils/model_arch_config_convertor.py#L147-L164" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/configs/model_config.py", "first": 1106, "last": 1111, "engine": "sglang", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L1106-L1111" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/config/model.py", "first": 1517, "last": 1537, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1517-L1537" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/model_executor/models/llama.py", "first": 140, "last": 158, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L140-L158" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 317, "last": 328, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L317-L328" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/layers/radix_attention.py", "first": 114, "last": 120, "engine": "sglang", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/radix_attention.py#L114-L120" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/v1/kv_cache_interface.py", "first": 233, "last": 248, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L233-L248" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 648, "last": 655, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L648-L655" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 427, "last": 444, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L427-L444" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 1190, "last": 1204, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L1190-L1204" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/v1/kv_cache_interface.py", "first": 392, "last": 392, "engine": "vllm", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L392-L392" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/models/deepseek_v2.py", "first": 1896, "last": 1928, "engine": "sglang", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_v2.py#L1896-L1928" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/models/deepseek_v2.py", "first": 1773, "last": 1773, "engine": "sglang", "section": "in-code", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_v2.py#L1773-L1773" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 326, "last": 333, "engine": "sglang", "section": "tp-asymmetry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L326-L333" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 248, "last": 265, "engine": "sglang", "section": "tp-asymmetry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L248-L265" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/server_args.py", "first": 1163, "last": 1170, "engine": "sglang", "section": "tp-asymmetry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1163-L1170" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/v1/kv_cache_interface.py", "first": 300, "last": 305, "engine": "vllm", "section": "tp-asymmetry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L300-L305" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "model_arch_config_convertor.py", "first": 147, "last": 164, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/transformers_utils/model_arch_config_convertor.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 764, "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "section": "in-code", "engine": "vllm", "path": "vllm/transformers_utils/model_arch_config_convertor.py", "first": 147, "last": 164 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/transformers_utils/model_arch_config_convertor.py#L147-L164" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "model_arch_config_convertor.py", "first": 99, "last": 113, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/transformers_utils/model_arch_config_convertor.py", "reviewed_symbol": "config", "resolved_engine": "vllm", "resolved_path": "vllm/transformers_utils/model_arch_config_convertor.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/transformers_utils/model_arch_config_convertor.py", "symbol": "config", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/transformers_utils/model_arch_config_convertor.py#L99-L113" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "llama.py", "first": 140, "last": 158, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/models/llama.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 767, "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "section": "in-code", "engine": "vllm", "path": "vllm/model_executor/models/llama.py", "first": 140, "last": 158 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L140-L158" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "attention.py", "first": 317, "last": 328, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/attention/attention.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 768, "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "section": "in-code", "engine": "vllm", "path": "vllm/model_executor/layers/attention/attention.py", "first": 317, "last": 328 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L317-L328" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "attention.py", "first": 648, "last": 655, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/attention/attention.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 771, "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "section": "in-code", "engine": "vllm", "path": "vllm/model_executor/layers/attention/attention.py", "first": 648, "last": 655 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L648-L655" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "kv_cache_interface.py", "first": 245, "last": 248, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/kv_cache_interface.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 770, "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "section": "in-code", "engine": "vllm", "path": "vllm/v1/kv_cache_interface.py", "first": 233, "last": 248 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L245-L248" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1038, "last": 1040, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1038-L1040" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1925, "last": 1931, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1925-L1931" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/config/model.py", "first": 1436, "last": 1442, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1436-L1442" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/config/model.py", "first": 1432, "last": 1442, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1432-L1442" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "kv_cache_utils.py", "first": 1925, "last": 1925, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/kv_cache_utils.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 788, "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "section": "worked-trace", "engine": "vllm", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1925, "last": 1931 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1925-L1925" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1096, "last": 1096, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1096-L1096" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "llama.py", "first": 145, "last": 152, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/models/llama.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 767, "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "section": "in-code", "engine": "vllm", "path": "vllm/model_executor/models/llama.py", "first": 140, "last": 158 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L145-L152" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 257, "last": 265, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L257-L265" }, { "page": "03-attention-kernels/03-05-mha-mqa-gqa-mla.html", "path": "vllm/v1/kv_cache_interface.py", "first": 300, "last": 305, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L300-L305" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/config/model.py", "first": 2496, "last": 2509, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L2496-L2509" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1592, "last": 1597, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1592-L1597" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/base.py", "first": 80, "last": 103, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/base.py#L80-L103" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/common.py", "first": 169, "last": 181, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/common.py#L169-L181" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2248, "last": 2251, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2248-L2251" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/models/llama.py", "first": 221, "last": 231, "engine": "vllm", "section": "rope-and-the-cache", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L221-L231" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "base.py", "first": 242, "last": 252, "engine": null, "section": "rope-and-the-cache", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/base.py", "reviewed_symbol": "rotary_embedding", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/base.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/base.py", "symbol": "rotary_embedding", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/base.py#L242-L252" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "linear_scaling_rope.py", "first": 90, "last": 97, "engine": null, "section": "context-extension", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py", "reviewed_symbol": "max_len", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py", "symbol": "max_len", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py#L90-L97" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "ntk_scaling_rope.py", "first": 31, "last": 36, "engine": null, "section": "context-extension", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py", "reviewed_symbol": "_compute_inv_freq", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py", "symbol": "_compute_inv_freq", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py#L31-L36" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/dynamic_ntk_scaling_rope.py", "first": 58, "last": 66, "engine": "vllm", "section": "context-extension", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/dynamic_ntk_scaling_rope.py#L58-L66" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "llama3_rope.py", "first": 33, "last": 54, "engine": null, "section": "context-extension", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/llama3_rope.py", "reviewed_symbol": "_compute_inv_freq", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/llama3_rope.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/llama3_rope.py", "symbol": "_compute_inv_freq", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/llama3_rope.py#L33-L54" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/common.py", "first": 34, "last": 76, "engine": "vllm", "section": "context-extension", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/common.py#L34-L76" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/yarn_scaling_rope.py", "first": 49, "last": 84, "engine": "vllm", "section": "context-extension", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/yarn_scaling_rope.py#L49-L84" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/__init__.py", "first": 64, "last": 65, "engine": "vllm", "section": "production-config", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/__init__.py#L64-L65" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "python/sglang/srt/layers/rotary_embedding/factory.py", "first": 34, "last": 51, "engine": "sglang", "section": "production-config", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/rotary_embedding/factory.py#L34-L51" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "python/sglang/srt/layers/rotary_embedding/base.py", "first": 184, "last": 196, "engine": "sglang", "section": "production-config", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/rotary_embedding/base.py#L184-L196" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/config/model.py", "first": 2446, "last": 2466, "engine": "vllm", "section": "production-config", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L2446-L2466" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 114, "last": 114, "engine": "sglang", "section": "sliding-window", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L114-L114" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "registry.py", "first": 159, "last": 167, "engine": null, "section": "sliding-window", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/registry.py", "reviewed_symbol": "UnifiedRadixCache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "symbol": "UnifiedRadixCache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L159-L167" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/kv_cache_interface.py", "first": 565, "last": 572, "engine": "vllm", "section": "sliding-window", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L565-L572" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "kv_cache_interface.py", "first": 300, "last": 306, "engine": null, "section": "sliding-window", "declared_engine": "vllm", "declared_path": "vllm/v1/kv_cache_interface.py", "reviewed_symbol": "max_memory_usage_bytes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/kv_cache_interface.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/kv_cache_interface.py", "symbol": "max_memory_usage_bytes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L300-L306" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 612, "last": 622, "engine": "vllm", "section": "sliding-window", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L612-L622" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 1095, "last": 1098, "engine": "vllm", "section": "sliding-window", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L1095-L1098" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "block_pool.py", "first": 187, "last": 191, "engine": null, "section": "sliding-window", "declared_engine": "vllm", "declared_path": "vllm/v1/core/block_pool.py", "reviewed_symbol": "represent", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/block_pool.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/block_pool.py", "symbol": "represent", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L187-L191" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "python/sglang/srt/configs/model_config.py", "first": 2114, "last": 2128, "engine": "sglang", "section": "sliding-window", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L2114-L2128" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "kv_cache_utils.py", "first": 1578, "last": 1600, "engine": null, "section": "sliding-window", "declared_engine": "vllm", "declared_path": "vllm/v1/core/kv_cache_utils.py", "reviewed_symbol": "function", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/kv_cache_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/kv_cache_utils.py", "symbol": "function", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1578-L1600" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "python/sglang/srt/mem_cache/swa_radix_cache.py", "first": 408, "last": 410, "engine": "sglang", "section": "sliding-window", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/swa_radix_cache.py#L408-L410" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "swa_radix_cache.py", "first": 70, "last": 74, "engine": null, "section": "sliding-window", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/swa_radix_cache.py", "reviewed_symbol": "invariant", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/swa_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/swa_radix_cache.py", "symbol": "invariant", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/swa_radix_cache.py#L70-L74" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "swa_radix_cache.py", "first": 640, "last": 672, "engine": null, "section": "sliding-window", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/swa_radix_cache.py", "reviewed_symbol": "swa_num_evicted", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/swa_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/swa_radix_cache.py", "symbol": "swa_num_evicted", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/swa_radix_cache.py#L640-L672" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "pure_swa_radix_cache.py", "first": 24, "last": 30, "engine": null, "section": "sliding-window", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/pure_swa_radix_cache.py", "reviewed_symbol": "PureSWARadixCache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/pure_swa_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/pure_swa_radix_cache.py", "symbol": "PureSWARadixCache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/pure_swa_radix_cache.py#L24-L30" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 1846, "last": 1850, "engine": "vllm", "section": "attention-sinks", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L1846-L1850" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/attention/static_sink_attention.py", "first": 86, "last": 99, "engine": "vllm", "section": "attention-sinks", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/static_sink_attention.py#L86-L99" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "python/sglang/srt/configs/model_config.py", "first": 816, "last": 822, "engine": "sglang", "section": "attention-sinks", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L816-L822" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/attention/selector.py", "first": 76, "last": 80, "engine": "vllm", "section": "attention-sinks", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L76-L80" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "gpu_model_runner.py", "first": 2248, "last": 2251, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 800, "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "section": "first-principles", "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2248, "last": 2251 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2248-L2251" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "llama.py", "first": 228, "last": 228, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/models/llama.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 801, "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "section": "rope-and-the-cache", "engine": "vllm", "path": "vllm/model_executor/models/llama.py", "first": 221, "last": 231 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L228-L228" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "base.py", "first": 221, "last": 252, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/base.py", "reviewed_symbol": "forward_cuda", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/base.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/base.py", "symbol": "forward_cuda", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/base.py#L221-L252" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "single_type_kv_cache_manager.py", "first": 624, "last": 657, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/core/single_type_kv_cache_manager.py", "reviewed_symbol": "remove_skipped_blocks", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/single_type_kv_cache_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "symbol": "remove_skipped_blocks", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L624-L657" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "kv_cache_utils.py", "first": 1569, "last": 1573, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/core/kv_cache_utils.py", "reviewed_symbol": "logger", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/kv_cache_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/kv_cache_utils.py", "symbol": "logger", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1569-L1573" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "kv_cache_interface.py", "first": 308, "last": 317, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/kv_cache_interface.py", "reviewed_symbol": "merge_window_sizes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/kv_cache_interface.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/kv_cache_interface.py", "symbol": "merge_window_sizes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L308-L317" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "base.py", "first": 58, "last": 61, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/base.py", "reviewed_symbol": "init_cache", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/base.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/base.py", "symbol": "init_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/base.py#L58-L61" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "base.py", "first": 102, "last": 105, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/rotary_embedding/base.py", "reviewed_symbol": "_compute_cos_sin_cache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/rotary_embedding/base.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/rotary_embedding/base.py", "symbol": "_compute_cos_sin_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/rotary_embedding/base.py#L102-L105" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/common.py", "first": 34, "last": 59, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/common.py#L34-L59" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/model_executor/layers/rotary_embedding/__init__.py", "first": 33, "last": 95, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/__init__.py#L33-L95" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 612, "last": 622, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L612-L622" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "linear_scaling_rope.py", "first": 34, "last": 37, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py", "reviewed_symbol": "supports", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py", "symbol": "supports", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py#L34-L37" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "ntk_scaling_rope.py", "first": 11, "last": 12, "engine": null, "section": "further-reading", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py", "reviewed_symbol": "RotaryEmbedding", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py", "symbol": "RotaryEmbedding", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/ntk_scaling_rope.py#L11-L12" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "dynamic_ntk_scaling_rope.py", "first": 33, "last": 33, "engine": null, "section": "further-reading", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/rotary_embedding/dynamic_ntk_scaling_rope.py", "reviewed_symbol": "Credits", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/rotary_embedding/dynamic_ntk_scaling_rope.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/rotary_embedding/dynamic_ntk_scaling_rope.py", "symbol": "Credits", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/rotary_embedding/dynamic_ntk_scaling_rope.py#L33-L33" }, { "page": "03-attention-kernels/03-06-positional-encoding-and-long-context.html", "path": "vllm/v1/kv_cache_interface.py", "first": 537, "last": 594, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L537-L594" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "python/sglang/srt/configs/model_config.py", "first": 1608, "last": 1618, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L1608-L1618" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "python/sglang/srt/server_args.py", "first": 143, "last": 179, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L143-L179" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "python/sglang/srt/configs/model_config.py", "first": 1473, "last": 1496, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L1473-L1496" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/model_executor/layers/quantization/utils/quant_utils.py", "first": 140, "last": 184, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/quant_utils.py#L140-L184" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/model_executor/layers/quantization/utils/quant_utils.py", "first": 105, "last": 137, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/quant_utils.py#L105-L137" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "quant_utils.py", "first": 195, "last": 308, "engine": null, "section": "first-principles", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/utils/quant_utils.py", "reviewed_symbol": "ScaleDesc", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/utils/quant_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/utils/quant_utils.py", "symbol": "ScaleDesc", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/quant_utils.py#L195-L308" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "first": 183, "last": 199, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L183-L199" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "first": 483, "last": 498, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py#L483-L498" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "auto_awq.py", "first": 183, "last": 195, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 851, "page": "04-quantization/04-01-the-design-space.html", "section": "first-principles", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "first": 183, "last": 199 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L183-L195" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "auto_gptq.py", "first": 106, "last": 131, "engine": null, "section": "first-principles", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "reviewed_symbol": "is_sym", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "symbol": "is_sym", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L106-L131" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/model_executor/layers/quantization/fp8.py", "first": 269, "last": 292, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/fp8.py#L269-L292" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "fp8.py", "first": 115, "last": 132, "engine": null, "section": "first-principles", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/fp8.py", "reviewed_symbol": "activation_scheme", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/fp8.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/fp8.py", "symbol": "activation_scheme", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/fp8.py#L115-L132" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 12, "last": 47, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L12-L47" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "python/sglang/srt/layers/quantization/__init__.py", "first": 71, "last": 143, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/__init__.py#L71-L143" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/config/cache.py", "first": 19, "last": 37, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L19-L37" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "python/sglang/srt/server_args.py", "first": 690, "last": 713, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L690-L713" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "first": 1053, "last": 1061, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py#L1053-L1061" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/config/quantization.py", "first": 114, "last": 146, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/quantization.py#L114-L146" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "python/sglang/srt/configs/model_config.py", "first": 1546, "last": 1587, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L1546-L1587" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/config/model.py", "first": 1251, "last": 1288, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1251-L1288" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/config/model.py", "first": 1314, "last": 1324, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1314-L1324" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "python/sglang/srt/configs/model_config.py", "first": 1497, "last": 1508, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L1497-L1508" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "compressed_tensors.py", "first": 299, "last": 369, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "reviewed_symbol": "_quantization_scheme_map_from_config", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "symbol": "_quantization_scheme_map_from_config", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py#L299-L369" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "compressed_tensors.py", "first": 934, "last": 968, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "reviewed_symbol": "get_scheme_dict", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "symbol": "get_scheme_dict", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py#L934-L968" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "compressed_tensors.py", "first": 722, "last": 883, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "reviewed_symbol": "_get_scheme_from_parts", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "symbol": "_get_scheme_from_parts", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py#L722-L883" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "fp8.py", "first": 99, "last": 99, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/fp8.py", "reviewed_symbol": "ignored_layers", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/fp8.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/fp8.py", "symbol": "ignored_layers", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/fp8.py#L99-L99" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "auto_awq.py", "first": 189, "last": 189, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 851, "page": "04-quantization/04-01-the-design-space.html", "section": "first-principles", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "first": 183, "last": 199 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L189-L189" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "modelopt.py", "first": 133, "last": 133, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/modelopt.py", "reviewed_symbol": "exclude_modules", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/modelopt.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/modelopt.py", "symbol": "exclude_modules", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/modelopt.py#L133-L133" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "w8a8_int8.py", "first": 76, "last": 76, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/quantization/w8a8_int8.py", "reviewed_symbol": "ignore", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/quantization/w8a8_int8.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/quantization/w8a8_int8.py", "symbol": "ignore", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/w8a8_int8.py#L76-L76" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/config/quantization.py", "first": 152, "last": 177, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/quantization.py#L152-L177" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/engine/arg_utils.py", "first": 802, "last": 806, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L802-L806" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "base_config.py", "first": 117, "last": 124, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/base_config.py", "reviewed_symbol": "get_min_capability", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/base_config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/base_config.py", "symbol": "get_min_capability", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/base_config.py#L117-L124" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "compressed_tensors.py", "first": 930, "last": 932, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "reviewed_symbol": "_check_scheme_supported", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", "symbol": "_check_scheme_supported", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py#L930-L932" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/config/model.py", "first": 1245, "last": 1346, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1245-L1346" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "vllm/config/quantization.py", "first": 158, "last": 194, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/quantization.py#L158-L194" }, { "page": "04-quantization/04-01-the-design-space.html", "path": "__init__.py", "first": 40, "last": 45, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/__init__.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 857, "page": "04-quantization/04-01-the-design-space.html", "section": "inventory", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 12, "last": 47 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L40-L45" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "first": 380, "last": 400, "engine": "vllm", "section": "gptq", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L380-L400" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "first": 436, "last": 457, "engine": "vllm", "section": "gptq", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L436-L457" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/llm_compressor/int4.md", "first": 136, "last": 138, "engine": null, "section": "gptq", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/llm_compressor/int4.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/llm_compressor/int4.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/llm_compressor/int4.md#L136-L138" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_wNa16.py", "first": 63, "last": 69, "engine": "vllm", "section": "gptq", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_wNa16.py#L63-L69" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_int8.py", "first": 57, "last": 86, "engine": "vllm", "section": "smoothquant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a8_int8.py#L57-L86" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "python/sglang/srt/layers/quantization/w8a8_int8.py", "first": 220, "last": 235, "engine": "sglang", "section": "smoothquant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/w8a8_int8.py#L220-L235" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/llm_compressor/int8_w8a8.md", "first": 85, "last": 94, "engine": null, "section": "smoothquant", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/llm_compressor/int8_w8a8.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/llm_compressor/int8_w8a8.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/llm_compressor/int8_w8a8.md#L85-L94" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/llm_compressor/int4.md", "first": 58, "last": 74, "engine": null, "section": "calibration", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/llm_compressor/int4.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/llm_compressor/int4.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/llm_compressor/int4.md#L58-L74" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/llm_compressor/int4.md", "first": 131, "last": 135, "engine": null, "section": "calibration", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/llm_compressor/int4.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/llm_compressor/int4.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/llm_compressor/int4.md#L131-L135" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/utils/quant_utils.py", "first": 561, "last": 580, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/quant_utils.py#L561-L580" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/_custom_ops.py", "first": 491, "last": 505, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/_custom_ops.py#L491-L505" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/awq_triton.py", "first": 48, "last": 68, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/awq_triton.py#L48-L68" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 139, "last": 152, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L139-L152" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "python/sglang/srt/layers/quantization/__init__.py", "first": 83, "last": 88, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/__init__.py#L83-L88" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "first": 919, "last": 937, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L919-L937" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "python/sglang/srt/hardware_backend/gpu/quantization/awq_kernels.py", "first": 94, "last": 105, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/hardware_backend/gpu/quantization/awq_kernels.py#L94-L105" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "python/sglang/srt/layers/quantization/gptq/schemes/gptq_linear.py", "first": 67, "last": 78, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/gptq/schemes/gptq_linear.py#L67-L78" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "auto_awq.py", "first": 238, "last": 253, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_awq.py", "reviewed_symbol": "from_config", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "symbol": "from_config", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L238-L253" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "auto_awq.py", "first": 285, "last": 331, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_awq.py", "reviewed_symbol": "get_quant_method", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "symbol": "get_quant_method", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L285-L331" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "auto_awq.py", "first": 453, "last": 491, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_awq.py", "reviewed_symbol": "PackedvLLMParameter", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "symbol": "PackedvLLMParameter", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L453-L491" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "auto_awq.py", "first": 93, "last": 168, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_awq.py", "reviewed_symbol": "_convert_awq_to_standard_format", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "symbol": "_convert_awq_to_standard_format", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L93-L168" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "first": 113, "last": 127, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L113-L127" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "first": 128, "last": 146, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/marlin.py#L128-L146" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "first": 229, "last": 250, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/marlin.py#L229-L250" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "first": 117, "last": 121, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L117-L121" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "python/sglang/srt/layers/quantization/gptq/schemes/gptq_linear.py", "first": 55, "last": 60, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/gptq/schemes/gptq_linear.py#L55-L60" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/llm_compressor/int4.md", "first": 126, "last": 127, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/llm_compressor/int4.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/llm_compressor/int4.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/llm_compressor/int4.md#L126-L127" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/llm_compressor/int4.md", "first": 135, "last": 135, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/llm_compressor/int4.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/llm_compressor/int4.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/llm_compressor/int4.md#L135-L135" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "auto_awq.py", "first": 448, "last": 450, "engine": null, "section": "hands-on", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_awq.py", "reviewed_symbol": "AutoAWQMarlinLinearMethod", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "symbol": "AutoAWQMarlinLinearMethod", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L448-L450" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/llm_compressor/int4.md", "first": 117, "last": 123, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/llm_compressor/int4.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/llm_compressor/int4.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/llm_compressor/int4.md#L117-L123" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "first": 366, "last": 378, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L366-L378" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "auto_awq.py", "first": 930, "last": 934, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 895, "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "section": "production", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "first": 919, "last": 937 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L930-L934" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "utils/marlin_utils.py", "first": 440, "last": 446, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 882, "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "section": "gptq", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "first": 436, "last": 457 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L440-L446" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "auto_gptq.py", "first": 412, "last": 417, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "reviewed_symbol": "scales_and_zp_size", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "symbol": "scales_and_zp_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L412-L417" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "awq_kernels.py", "first": 100, "last": 101, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/hardware_backend/gpu/quantization/awq_kernels.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 896, "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "section": "production", "engine": "sglang", "path": "python/sglang/srt/hardware_backend/gpu/quantization/awq_kernels.py", "first": 94, "last": 105 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/hardware_backend/gpu/quantization/awq_kernels.py#L100-L101" }, { "page": "04-quantization/04-02-gptq-awq-smoothquant.html", "path": "docs/features/quantization/auto_awq.md", "first": 3, "last": 5, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/quantization/auto_awq.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/quantization/auto_awq.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/quantization/auto_awq.md#L3-L5" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "first": 68, "last": 93, "engine": "vllm", "section": "what-the-tensor-cores-provide", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_mma.h#L68-L93" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_mma.h", "first": 23, "last": 133, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "reviewed_symbol": "m16n8k32", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "symbol": "m16n8k32", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_mma.h#L23-L133" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_mma.h", "first": 39, "last": 39, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "reviewed_symbol": "mma.sync.aligned.m16n8k16", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "symbol": "mma.sync.aligned.m16n8k16", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_mma.h#L39-L39" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_mma.h", "first": 87, "last": 87, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 917, "page": "04-quantization/04-03-formats-and-kernels.html", "section": "what-the-tensor-cores-provide", "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "first": 68, "last": 93 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_mma.h#L87-L87" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_mma.h", "first": 79, "last": 79, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 917, "page": "04-quantization/04-03-formats-and-kernels.html", "section": "what-the-tensor-cores-provide", "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin_mma.h", "first": 68, "last": 93 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_mma.h#L79-L79" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "machete_prepacked_layout.cuh", "first": 100, "last": 103, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "reviewed_symbol": "rs_op_selector", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "symbol": "rs_op_selector", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh#L100-L103" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "nvfp4_scaled_mm_kernels.cu", "first": 77, "last": 98, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu", "reviewed_symbol": "OpClassBlockScaledTensorOp", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu", "symbol": "OpClassBlockScaledTensorOp", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu#L77-L98" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "dequant.h", "first": 107, "last": 119, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 926, "page": "04-quantization/04-03-formats-and-kernels.html", "section": "dequant-inside-the-pipeline", "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "first": 107, "last": 119 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/dequant.h#L107-L119" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "nvfp4_scaled_mm_kernels.cu", "first": 155, "last": 168, "engine": null, "section": "what-the-tensor-cores-provide", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu", "reviewed_symbol": "tile_atom_to_shape_SFA", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu", "symbol": "tile_atom_to_shape_SFA", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/fp4/nvfp4_scaled_mm_kernels.cu#L155-L168" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "first": 107, "last": 119, "engine": "vllm", "section": "dequant-inside-the-pipeline", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/dequant.h#L107-L119" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "dequant.h", "first": 121, "last": 143, "engine": null, "section": "dequant-inside-the-pipeline", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "reviewed_symbol": "__hsub2", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "symbol": "__hsub2", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/dequant.h#L121-L143" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "first": 104, "last": 116, "engine": "vllm", "section": "dequant-inside-the-pipeline", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_template.h#L104-L116" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_template.h", "first": 19, "last": 19, "engine": null, "section": "marlin", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "reviewed_symbol": "IST-DASLab/marlin", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "symbol": "IST-DASLab/marlin", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_template.h#L19-L19" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/gptq_marlin_repack.cu", "first": 122, "last": 134, "engine": "vllm", "section": "marlin", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/gptq_marlin_repack.cu#L122-L134" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/gptq_marlin_repack.cu", "first": 199, "last": 209, "engine": "vllm", "section": "marlin", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/gptq_marlin_repack.cu#L199-L209" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin.cuh", "first": 21, "last": 31, "engine": "vllm", "section": "marlin", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin.cuh#L21-L31" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin.cu", "first": 193, "last": 226, "engine": null, "section": "marlin", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "reviewed_symbol": "sh_b_size", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "symbol": "sh_b_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin.cu#L193-L226" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin.cu", "first": 229, "last": 260, "engine": null, "section": "marlin", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "reviewed_symbol": "is_valid_config", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "symbol": "is_valid_config", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin.cu#L229-L260" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "first": 1795, "last": 1816, "engine": "vllm", "section": "marlin", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_template.h#L1795-L1816" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_template.h", "first": 924, "last": 931, "engine": null, "section": "marlin", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "reviewed_symbol": "wait_for_stage", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "symbol": "wait_for_stage", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_template.h#L924-L931" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "first": 139, "last": 153, "engine": "vllm", "section": "marlin", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin.cu#L139-L153" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin.cu", "first": 276, "last": 325, "engine": null, "section": "marlin", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "reviewed_symbol": "determine_exec_config", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin.cu", "symbol": "determine_exec_config", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin.cu#L276-L325" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "first": 221, "last": 244, "engine": "vllm", "section": "marlin", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L221-L244" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "mixed_precision/marlin.py", "first": 62, "last": 69, "engine": null, "section": "marlin", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "reviewed_symbol": "check_marlin_supports_shape", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "symbol": "check_marlin_supports_shape", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/marlin.py#L62-L69" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "first": 179, "last": 196, "engine": "vllm", "section": "marlin", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L179-L196" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_utils.py", "first": 36, "last": 36, "engine": null, "section": "marlin", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "reviewed_symbol": "MARLIN_SUPPORTED_GROUP_SIZES", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "symbol": "MARLIN_SUPPORTED_GROUP_SIZES", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L36-L36" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/machete/Readme.md", "first": 1, "last": 3, "engine": "vllm", "section": "machete", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/Readme.md#L1-L3" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "first": 71, "last": 80, "engine": "vllm", "section": "machete", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh#L71-L80" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "machete_prepacked_layout.cuh", "first": 100, "last": 103, "engine": null, "section": "machete", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "reviewed_symbol": "rs_op_selector", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "symbol": "rs_op_selector", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh#L100-L103" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "machete_prepacked_layout.cuh", "first": 81, "last": 86, "engine": null, "section": "machete", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "reviewed_symbol": "Ct = Bt * At", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "symbol": "Ct = Bt * At", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh#L81-L86" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "mixed_precision/marlin.py", "first": 36, "last": 38, "engine": null, "section": "machete", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "reviewed_symbol": "get_min_capability", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "symbol": "get_min_capability", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/marlin.py#L36-L38" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "mixed_precision/machete.py", "first": 35, "last": 36, "engine": null, "section": "machete", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "reviewed_symbol": "is_device_capability", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "symbol": "is_device_capability", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/machete.py#L35-L36" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/layers/quantization/utils/machete_utils.py", "first": 37, "last": 51, "engine": "vllm", "section": "machete", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/machete_utils.py#L37-L51" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/cutlass_extensions/epilogue/scaled_mm_epilogues_c3x.hpp", "first": 150, "last": 173, "engine": "vllm", "section": "fp8-and-cutlass", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/cutlass_extensions/epilogue/scaled_mm_epilogues_c3x.hpp#L150-L173" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "scaled_mm/cutlass.py", "first": 58, "last": 69, "engine": null, "section": "fp8-and-cutlass", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/scaled_mm/cutlass.py", "reviewed_symbol": "convert_to_channelwise", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/scaled_mm/cutlass.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/scaled_mm/cutlass.py", "symbol": "convert_to_channelwise", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/scaled_mm/cutlass.py#L58-L69" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "scaled_mm_epilogues_c3x.hpp", "first": 141, "last": 141, "engine": null, "section": "fp8-and-cutlass", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/cutlass_extensions/epilogue/scaled_mm_epilogues_c3x.hpp", "reviewed_symbol": "symmetric quantization", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/cutlass_extensions/epilogue/scaled_mm_epilogues_c3x.hpp", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/cutlass_extensions/epilogue/scaled_mm_epilogues_c3x.hpp", "symbol": "symmetric quantization", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/cutlass_extensions/epilogue/scaled_mm_epilogues_c3x.hpp#L141-L141" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 139, "last": 153, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L139-L153" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "first": 340, "last": 358, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L340-L358" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "auto_awq.py", "first": 430, "last": 448, "engine": null, "section": "kernel-selection", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_awq.py", "reviewed_symbol": "MPLinearLayerConfig", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_awq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_awq.py", "symbol": "MPLinearLayerConfig", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_awq.py#L430-L448" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/kernels/linear/__init__.py", "first": 353, "last": 383, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L353-L383" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "auto_gptq.py", "first": 320, "last": 324, "engine": null, "section": "kernel-selection", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "reviewed_symbol": "verify_marlin_supported", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "symbol": "verify_marlin_supported", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L320-L324" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/kernels/linear/base.py", "first": 168, "last": 210, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/base.py#L168-L210" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/kernels/linear/__init__.py", "first": 474, "last": 484, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L474-L484" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "kernels/linear/__init__.py", "first": 826, "last": 837, "engine": null, "section": "kernel-selection", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/__init__.py", "reviewed_symbol": "failure_reasons", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/__init__.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/__init__.py", "symbol": "failure_reasons", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L826-L837" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "kernels/linear/__init__.py", "first": 809, "last": 813, "engine": null, "section": "kernel-selection", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/__init__.py", "reviewed_symbol": "VLLM_DISABLED_KERNELS", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/__init__.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/__init__.py", "symbol": "VLLM_DISABLED_KERNELS", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L809-L813" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh", "first": 322, "last": 349, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh#L322-L349" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/machete/generate.py", "first": 509, "last": 537, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/generate.py#L509-L537" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/kernels/linear/mxfp8/rocm_native.py", "first": 122, "last": 145, "engine": "vllm", "section": "kernel-selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mxfp8/rocm_native.py#L122-L145" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "python/sglang/srt/layers/quantization/marlin_utils.py", "first": 3, "last": 3, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/marlin_utils.py#L3-L3" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "python/sglang/kernels/ops/quantization/gptq_marlin.py", "first": 18, "last": 26, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/quantization/gptq_marlin.py#L18-L26" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "python/sglang/srt/layers/quantization/utils.py", "first": 508, "last": 508, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/utils.py#L508-L508" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 12, "last": 46, "engine": "vllm", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L12-L46" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "python/sglang/srt/layers/quantization/fp8_utils.py", "first": 774, "last": 792, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/fp8_utils.py#L774-L792" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "python/sglang/srt/layers/deep_gemm_wrapper/entrypoint.py", "first": 189, "last": 193, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/deep_gemm_wrapper/entrypoint.py#L189-L193" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "configurer.py", "first": 21, "last": 23, "engine": null, "section": "sglang", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/deep_gemm_wrapper/configurer.py", "reviewed_symbol": "sm_version", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/deep_gemm_wrapper/configurer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/deep_gemm_wrapper/configurer.py", "symbol": "sm_version", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/deep_gemm_wrapper/configurer.py#L21-L23" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "deep_gemm_wrapper/configurer.py", "first": 27, "last": 31, "engine": null, "section": "sglang", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/deep_gemm_wrapper/configurer.py", "reviewed_symbol": "deep_gemm", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/deep_gemm_wrapper/configurer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/deep_gemm_wrapper/configurer.py", "symbol": "deep_gemm", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/deep_gemm_wrapper/configurer.py#L27-L31" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "python/sglang/srt/layers/quantization/fp8_utils.py", "first": 915, "last": 934, "engine": "sglang", "section": "sglang", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/fp8_utils.py#L915-L934" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "auto_gptq.py", "first": 340, "last": 358, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 954, "page": "04-quantization/04-03-formats-and-kernels.html", "section": "kernel-selection", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "first": 340, "last": 358 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L340-L358" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "kernels/linear/__init__.py", "first": 775, "last": 837, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/__init__.py", "reviewed_symbol": "choose_mp_linear_kernel", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/__init__.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/__init__.py", "symbol": "choose_mp_linear_kernel", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L775-L837" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "mixed_precision/machete.py", "first": 30, "last": 64, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "reviewed_symbol": "can_implement", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "symbol": "can_implement", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/machete.py#L30-L64" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "machete.py", "first": 70, "last": 104, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "reviewed_symbol": "machete_prepack_B", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mixed_precision/machete.py", "symbol": "machete_prepack_B", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/machete.py#L70-L104" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "generate.py", "first": 45, "last": 57, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/machete/generate.py", "reviewed_symbol": "mm_dispatch_", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/generate.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/machete/generate.py", "symbol": "mm_dispatch_", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/generate.py#L45-L57" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "machete_prepacked_layout.cuh", "first": 56, "last": 59, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "reviewed_symbol": "should_interleave", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh", "symbol": "should_interleave", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/machete_prepacked_layout.cuh#L56-L59" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "auto_gptq.py", "first": 356, "last": 358, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/auto_gptq.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 954, "page": "04-quantization/04-03-formats-and-kernels.html", "section": "kernel-selection", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "first": 340, "last": 358 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L356-L358" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_utils.py", "first": 179, "last": 196, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 941, "page": "04-quantization/04-03-formats-and-kernels.html", "section": "marlin", "engine": "vllm", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "first": 179, "last": 196 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L179-L196" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "mixed_precision/marlin.py", "first": 62, "last": 69, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "reviewed_symbol": "check_marlin_supports_shape", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "symbol": "check_marlin_supports_shape", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/marlin.py#L62-L69" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "mixed_precision/marlin.py", "first": 76, "last": 81, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "reviewed_symbol": "group_size", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mixed_precision/marlin.py", "symbol": "group_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mixed_precision/marlin.py#L76-L81" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_utils.py", "first": 633, "last": 654, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "reviewed_symbol": "should_use_atomic_add_reduce", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "symbol": "should_use_atomic_add_reduce", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L633-L654" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "rocm_native.py", "first": 165, "last": 170, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/kernels/linear/mxfp8/rocm_native.py", "reviewed_symbol": "BLOCK_M", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/kernels/linear/mxfp8/rocm_native.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/kernels/linear/mxfp8/rocm_native.py", "symbol": "BLOCK_M", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/mxfp8/rocm_native.py#L165-L170" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "dequant.h", "first": 226, "last": 250, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "reviewed_symbol": "kU8B128", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "symbol": "kU8B128", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/dequant.h#L226-L250" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "generate.py", "first": 530, "last": 532, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/generate.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 963, "page": "04-quantization/04-03-formats-and-kernels.html", "section": "kernel-selection", "engine": "vllm", "path": "csrc/libtorch_stable/quantization/machete/generate.py", "first": 509, "last": 537 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/generate.py#L530-L532" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "marlin_template.h", "first": 19, "last": 19, "engine": null, "section": "further-reading", "declared_engine": "vllm", "declared_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "reviewed_symbol": "IST-DASLab/marlin", "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "symbol": "IST-DASLab/marlin", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_template.h#L19-L19" }, { "page": "04-quantization/04-03-formats-and-kernels.html", "path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "first": 1, "last": 64, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/dequant.h#L1-L64" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "csrc/libtorch_stable/quantization/marlin/dequant.h", "first": 173, "last": 199, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/dequant.h#L173-L199" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "first": 104, "last": 116, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_template.h#L104-L116" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "first": 348, "last": 354, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/quantization/marlin/marlin_template.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/marlin/marlin_template.h#L348-L354" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/config/cache.py", "first": 80, "last": 87, "engine": "vllm", "section": "capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L80-L87" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/config/quantization.py", "first": 138, "last": 147, "engine": "vllm", "section": "moe", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/quantization.py#L138-L147" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/config/quantization.py", "first": 65, "last": 77, "engine": "vllm", "section": "moe", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/quantization.py#L65-L77" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "python/sglang/srt/layers/quantization/w4afp8.py", "first": 93, "last": 105, "engine": "sglang", "section": "moe", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/w4afp8.py#L93-L105" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "python/sglang/srt/layers/quantization/__init__.py", "first": 71, "last": 101, "engine": "sglang", "section": "moe", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/__init__.py#L71-L101" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "tests/evals/gsm8k/configs/Qwen3-4B-TQ-k8v4.yaml", "first": 1, "last": 5, "engine": "vllm", "section": "accuracy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/configs/Qwen3-4B-TQ-k8v4.yaml#L1-L5" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "tests/evals/gsm8k/configs/Qwen3-4B-TQ-t3nc.yaml", "first": 1, "last": 5, "engine": "vllm", "section": "accuracy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/configs/Qwen3-4B-TQ-t3nc.yaml#L1-L5" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "tests/evals/gsm8k/test_gsm8k_correctness.py", "first": 176, "last": 192, "engine": "vllm", "section": "accuracy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/test_gsm8k_correctness.py#L176-L192" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "tests/evals/gsm8k/gsm8k_eval.py", "first": 210, "last": 223, "engine": null, "section": "accuracy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/evals/gsm8k/gsm8k_eval.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/evals/gsm8k/gsm8k_eval.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/gsm8k_eval.py#L210-L223" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "tests/evals/gsm8k/gsm8k_eval.py", "first": 68, "last": 77, "engine": null, "section": "accuracy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/evals/gsm8k/gsm8k_eval.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/evals/gsm8k/gsm8k_eval.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/gsm8k_eval.py#L68-L77" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 150, "last": 152, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L150-L152" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/model_executor/kernels/linear/scaled_mm/marlin.py", "first": 29, "last": 45, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/scaled_mm/marlin.py#L29-L45" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/model_executor/kernels/linear/scaled_mm/flashinfer.py", "first": 304, "last": 316, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/scaled_mm/flashinfer.py#L304-L316" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/benchmarks/latency.py", "first": 34, "last": 52, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L34-L52" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/benchmarks/latency.py", "first": 125, "last": 167, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L125-L167" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "python/sglang/benchmark/one_batch.py", "first": 784, "last": 807, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L784-L807" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "python/sglang/benchmark/one_batch.py", "first": 855, "last": 865, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L855-L865" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "python/sglang/benchmark/one_batch.py", "first": 215, "last": 226, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L215-L226" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "first": 178, "last": 186, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L178-L186" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 150, "last": 152, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L150-L152" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/model_executor/layers/quantization/auto_gptq.py", "first": 219, "last": 238, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/auto_gptq.py#L219-L238" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "python/sglang/srt/layers/quantization/__init__.py", "first": 86, "last": 88, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/__init__.py#L86-L88" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/model_executor/layers/quantization/utils/marlin_utils.py", "first": 221, "last": 242, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/utils/marlin_utils.py#L221-L242" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "benchmarks/benchmark_latency.py", "first": 1, "last": 17, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "benchmarks/benchmark_latency.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "benchmarks/benchmark_latency.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_latency.py#L1-L17" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "python/sglang/bench_one_batch.py", "first": 1, "last": 19, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/bench_one_batch.py#L1-L19" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/benchmarks/serve.py", "first": 1588, "last": 1599, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1588-L1599" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "tests/evals/gsm8k/gsm8k_eval.py", "first": 189, "last": 200, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/evals/gsm8k/gsm8k_eval.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/evals/gsm8k/gsm8k_eval.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/gsm8k_eval.py#L189-L200" }, { "page": "04-quantization/04-04-when-quantization-pays.html", "path": "vllm/config/quantization.py", "first": 138, "last": 147, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/quantization.py#L138-L147" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 1656, "last": 1656, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1656-L1656" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/layers/linear.py", "first": 1615, "last": 1615, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/linear.py#L1615-L1615" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/models/llama.py", "first": 140, "last": 153, "engine": "vllm", "section": "heads", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L140-L153" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/models/llama.py", "first": 158, "last": 171, "engine": "sglang", "section": "heads", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/llama.py#L158-L171" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 1022, "last": 1030, "engine": "vllm", "section": "heads", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1022-L1030" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/models/llama.py", "first": 305, "last": 308, "engine": "vllm", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L305-L308" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/vocab_parallel_embedding.py", "first": 487, "last": 507, "engine": "vllm", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/vocab_parallel_embedding.py#L487-L507" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/layers/vocab_parallel_embedding.py", "first": 566, "last": 579, "engine": "sglang", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/vocab_parallel_embedding.py#L566-L579" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/logits_processor.py", "first": 171, "last": 187, "engine": "vllm", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/logits_processor.py#L171-L187" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/logits_processor.py", "first": 194, "last": 200, "engine": "vllm", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/logits_processor.py#L194-L200" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/models/llama.py", "first": 446, "last": 455, "engine": "vllm", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L446-L455" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/vocab_parallel_embedding.py", "first": 32, "last": 32, "engine": "vllm", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/vocab_parallel_embedding.py#L32-L32" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/layers/vocab_parallel_embedding.py", "first": 48, "last": 48, "engine": "sglang", "section": "not-split", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/vocab_parallel_embedding.py#L48-L48" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 454, "last": 473, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L454-L473" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 1653, "last": 1662, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1653-L1662" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/distributed/communication_op.py", "first": 12, "last": 14, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/communication_op.py#L12-L14" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/distributed/communication_op.py", "first": 12, "last": 14, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/communication_op.py#L12-L14" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/layers/linear.py", "first": 1638, "last": 1660, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/linear.py#L1638-L1660" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/compilation/passes/fusion/sequence_parallelism.py", "first": 498, "last": 515, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/passes/fusion/sequence_parallelism.py#L498-L515" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/config/vllm.py", "first": 303, "last": 309, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L303-L309" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/config/vllm.py", "first": 435, "last": 435, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L435-L435" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/compilation/passes/pass_manager.py", "first": 161, "last": 164, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/passes/pass_manager.py#L161-L164" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/compilation/passes/fusion/sequence_parallelism.py", "first": 604, "last": 618, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/passes/fusion/sequence_parallelism.py#L604-L618" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 548, "last": 557, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L548-L557" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 1614, "last": 1622, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1614-L1622" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 1271, "last": 1280, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1271-L1280" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/models/llama.py", "first": 140, "last": 153, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L140-L153" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 1022, "last": 1030, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1022-L1030" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 454, "last": 473, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L454-L473" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/models/llama.py", "first": 345, "last": 354, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L345-L354" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/models/llama.py", "first": 221, "last": 231, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L221-L231" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 1641, "last": 1668, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1641-L1668" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/configs/model_config.py", "first": 1166, "last": 1177, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L1166-L1177" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 326, "last": 333, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L326-L333" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/config/model.py", "first": 1517, "last": 1537, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1517-L1537" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/config/model.py", "first": 1411, "last": 1418, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1411-L1418" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 689, "last": 689, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L689-L689" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/distributed/utils.py", "first": 53, "last": 57, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/utils.py#L53-L57" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1925, "last": 1931, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1925-L1931" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/model_executor/layers/linear.py", "first": 595, "last": 601, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L595-L601" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 1025, "last": 1032, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1025-L1032" }, { "page": "05-parallelism/05-01-tensor-parallelism.html", "path": "vllm/config/parallel.py", "first": 124, "last": 125, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L124-L125" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/config/model.py", "first": 1423, "last": 1430, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1423-L1430" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/model_executor/models/llama.py", "first": 408, "last": 434, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L408-L434" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/models/llama.py", "first": 427, "last": 437, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/llama.py#L427-L437" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/distributed/parallel_state.py", "first": 1057, "last": 1064, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/parallel_state.py#L1057-L1064" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1134, "last": 1138, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1134-L1138" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "first": 1044, "last": 1052, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_pp_mixin.py#L1044-L1052" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/model_executor/model_runner_components/layer_setup.py", "first": 132, "last": 151, "engine": "sglang", "section": "kv-and-the-decision-rule", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner_components/layer_setup.py#L132-L151" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 150, "last": 160, "engine": "sglang", "section": "kv-and-the-decision-rule", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L150-L160" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/mem_cache/kv_cache_configurator.py", "first": 1123, "last": 1123, "engine": "sglang", "section": "kv-and-the-decision-rule", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_configurator.py#L1123-L1123" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 7896, "last": 7909, "engine": "vllm", "section": "kv-and-the-decision-rule", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L7896-L7909" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/distributed/utils.py", "first": 127, "last": 142, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/utils.py#L127-L142" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/model_executor/models/utils.py", "first": 819, "last": 862, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/utils.py#L819-L862" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/config/vllm.py", "first": 583, "last": 594, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L583-L594" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/v1/engine/core.py", "first": 205, "last": 215, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L205-L215" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "first": 69, "last": 92, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_pp_mixin.py#L69-L92" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "first": 561, "last": 573, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_pp_mixin.py#L561-L573" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 1054, "last": 1056, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1054-L1056" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "scheduler_pp_mixin.py", "first": 967, "last": 982, "engine": null, "section": "how-production-systems-do-it", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "reviewed_symbol": "_pp_send_pyobj_to_next_stage", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "symbol": "_pp_send_pyobj_to_next_stage", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_pp_mixin.py#L967-L982" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "first": 1056, "last": 1065, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_pp_mixin.py#L1056-L1065" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1059, "last": 1065, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1059-L1065" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/v1/engine/core.py", "first": 624, "last": 680, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L624-L680" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1098, "last": 1103, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1098-L1103" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "gpu_model_runner.py", "first": 3733, "last": 3739, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "intermediate_tensors", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "intermediate_tensors", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3733-L3739" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "llama.py", "first": 430, "last": 433, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/model_executor/models/llama.py", "reviewed_symbol": "IntermediateTensors", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/models/llama.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/models/llama.py", "symbol": "IntermediateTensors", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L430-L433" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "gpu_worker.py", "first": 1127, "last": 1138, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_worker.py", "reviewed_symbol": "isend_tensor_dict", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_worker.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_worker.py", "symbol": "isend_tensor_dict", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1127-L1138" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "parallel_state.py", "first": 1061, "last": 1064, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/distributed/parallel_state.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1066, "page": "05-parallelism/05-02-pipeline-parallelism.html", "section": "first-principles", "engine": "vllm", "path": "vllm/distributed/parallel_state.py", "first": 1057, "last": 1064 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/parallel_state.py#L1061-L1064" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "gpu_model_runner.py", "first": 3472, "last": 3500, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "sync_and_gather_intermediate_tensors", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "sync_and_gather_intermediate_tensors", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3472-L3500" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/managers/scheduler_pp_mixin.py", "first": 1230, "last": 1241, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_pp_mixin.py#L1230-L1241" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 9245, "last": 9253, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9245-L9253" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 4640, "last": 4640, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4640-L4640" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/distributed/utils.py", "first": 160, "last": 166, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/utils.py#L160-L166" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 159, "last": 165, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L159-L165" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 1065, "last": 1084, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L1065-L1084" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/distributed/utils.py", "first": 156, "last": 170, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/utils.py#L156-L170" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 9245, "last": 9248, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9245-L9248" }, { "page": "05-parallelism/05-02-pipeline-parallelism.html", "path": "vllm/config/vllm.py", "first": 588, "last": 594, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L588-L594" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/managers/data_parallel_controller.py", "first": 79, "last": 93, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/data_parallel_controller.py#L79-L93" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/managers/data_parallel_controller.py", "first": 114, "last": 130, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/data_parallel_controller.py#L114-L130" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/communicator.py", "first": 197, "last": 210, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/communicator.py#L197-L210" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/dp_attention.py", "first": 326, "last": 340, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/dp_attention.py#L326-L340" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 1163, "last": 1170, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1163-L1170" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/kernels/aot/benchmark/bench_fp8_blockwise_group_gemm.py", "first": 320, "last": 325, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/aot/benchmark/bench_fp8_blockwise_group_gemm.py#L320-L325" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/moe/token_dispatcher/deepep.py", "first": 527, "last": 534, "engine": "sglang", "section": "expert-parallelism", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/token_dispatcher/deepep.py#L527-L534" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/communicator.py", "first": 385, "last": 400, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/communicator.py#L385-L400" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/dp_attention.py", "first": 120, "last": 127, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/dp_attention.py#L120-L127" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/managers/scheduler_components/dp_attn.py", "first": 359, "last": 369, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/dp_attn.py#L359-L369" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/managers/scheduler_components/dp_attn.py", "first": 131, "last": 193, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/dp_attn.py#L131-L193" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/config/parallel.py", "first": 129, "last": 131, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L129-L131" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/model_executor/layers/fused_moe/config.py", "first": 1202, "last": 1210, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/config.py#L1202-L1210" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/model_executor/layers/fused_moe/config.py", "first": 1243, "last": 1253, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/config.py#L1243-L1253" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/distributed/device_communicators/all2all.py", "first": 44, "last": 48, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/all2all.py#L44-L48" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/v1/engine/coordinator.py", "first": 33, "last": 41, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/coordinator.py#L33-L41" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/v1/engine/core.py", "first": 2202, "last": 2206, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L2202-L2206" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/v1/engine/core.py", "first": 2253, "last": 2257, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L2253-L2257" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/config/parallel.py", "first": 149, "last": 155, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L149-L155" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/expert_distribution.py", "first": 1088, "last": 1103, "engine": "sglang", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/expert_distribution.py#L1088-L1103" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/distributed/eplb/eplb_state.py", "first": 600, "last": 612, "engine": "vllm", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/eplb/eplb_state.py#L600-L612" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/eplb_algorithms/deepseek.py", "first": 70, "last": 82, "engine": "sglang", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/eplb_algorithms/deepseek.py#L70-L82" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/eplb_algorithms/deepseek.py", "first": 126, "last": 151, "engine": "sglang", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/eplb_algorithms/deepseek.py#L126-L151" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/expert_location.py", "first": 237, "last": 241, "engine": "sglang", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/expert_location.py#L237-L241" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/config/parallel.py", "first": 72, "last": 73, "engine": "vllm", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L72-L73" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/eplb_manager.py", "first": 91, "last": 97, "engine": "sglang", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/eplb_manager.py#L91-L97" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/expert_location_updater.py", "first": 489, "last": 505, "engine": "sglang", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/expert_location_updater.py#L489-L505" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 2435, "last": 2450, "engine": "sglang", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2435-L2450" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/config/parallel.py", "first": 62, "last": 87, "engine": "vllm", "section": "eplb", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L62-L87" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/elastic_ep/elastic_ep.py", "first": 43, "last": 72, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/elastic_ep/elastic_ep.py#L43-L72" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/elastic_ep/elastic_ep.py", "first": 90, "last": 104, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/elastic_ep/elastic_ep.py#L90-L104" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/managers/data_parallel_controller.py", "first": 566, "last": 578, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/data_parallel_controller.py#L566-L578" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 2499, "last": 2518, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2499-L2518" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/elastic_ep/elastic_ep.py", "first": 166, "last": 242, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/elastic_ep/elastic_ep.py#L166-L242" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 2529, "last": 2534, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2529-L2534" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/elastic_ep/expert_backup_manager.py", "first": 39, "last": 52, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/elastic_ep/expert_backup_manager.py#L39-L52" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/eplb_manager.py", "first": 190, "last": 225, "engine": "sglang", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/eplb_manager.py#L190-L225" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/v1/engine/core.py", "first": 2272, "last": 2290, "engine": "vllm", "section": "elastic-ep", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L2272-L2290" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/managers/data_parallel_controller.py", "first": 114, "last": 130, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/data_parallel_controller.py#L114-L130" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/managers/scheduler_components/dp_attn.py", "first": 359, "last": 369, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/dp_attn.py#L359-L369" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/dp_attention.py", "first": 326, "last": 340, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/dp_attention.py#L326-L340" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/communicator.py", "first": 385, "last": 400, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/communicator.py#L385-L400" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/moe/token_dispatcher/deepep.py", "first": 766, "last": 776, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/token_dispatcher/deepep.py#L766-L776" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/moe/token_dispatcher/deepep.py", "first": 726, "last": 730, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/token_dispatcher/deepep.py#L726-L730" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 6742, "last": 6744, "engine": "sglang", "section": "combining", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L6742-L6744" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/model_executor/layers/fused_moe/config.py", "first": 1202, "last": 1210, "engine": "vllm", "section": "combining", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/config.py#L1202-L1210" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/kernels/aot/benchmark/bench_fp8_blockwise_group_gemm.py", "first": 326, "last": 333, "engine": "sglang", "section": "combining", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/aot/benchmark/bench_fp8_blockwise_group_gemm.py#L326-L333" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/server_args.py", "first": 6746, "last": 6746, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L6746-L6746" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/moe/token_dispatcher/deepep.py", "first": 389, "last": 397, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/token_dispatcher/deepep.py#L389-L397" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/config/parallel.py", "first": 507, "last": 528, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L507-L528" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/expert_location_updater.py", "first": 489, "last": 492, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/expert_location_updater.py#L489-L492" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "examples/features/data_parallel/data_parallel_offline.py", "first": 4, "last": 9, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "examples/features/data_parallel/data_parallel_offline.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "examples/features/data_parallel/data_parallel_offline.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/features/data_parallel/data_parallel_offline.py#L4-L9" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/layers/dp_attention.py", "first": 95, "last": 127, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/dp_attention.py#L95-L127" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "python/sglang/srt/eplb/eplb_algorithms/deepseek.py", "first": 70, "last": 82, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/eplb/eplb_algorithms/deepseek.py#L70-L82" }, { "page": "05-parallelism/05-03-data-and-expert-parallelism.html", "path": "vllm/model_executor/layers/fused_moe/config.py", "first": 1202, "last": 1210, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/config.py#L1202-L1210" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/device_communicators/custom_all_reduce.py", "first": 165, "last": 186, "engine": "vllm", "section": "topology", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/custom_all_reduce.py#L165-L186" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/platforms/cuda.py", "first": 789, "last": 804, "engine": "vllm", "section": "topology", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L789-L804" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_all_reduce.cuh", "first": 7, "last": 23, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_reduce.cuh#L7-L23" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_collective_common.cuh", "first": 202, "last": 217, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_collective_common.cuh#L202-L217" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_all_reduce.cuh", "first": 30, "last": 75, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/custom_all_reduce.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/custom_all_reduce.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_reduce.cuh#L30-L75" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_all_reduce.cuh", "first": 291, "last": 310, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_reduce.cuh#L291-L310" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_collective_common.cuh", "first": 34, "last": 39, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/custom_collective_common.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/custom_collective_common.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_collective_common.cuh#L34-L39" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/device_communicators/custom_all_reduce.py", "first": 348, "last": 361, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/custom_all_reduce.py#L348-L361" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/device_communicators/cuda_communicator.py", "first": 287, "last": 327, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/cuda_communicator.py#L287-L327" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/quickreduce/quick_reduce.h", "first": 208, "last": 230, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/quickreduce/quick_reduce.h", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/quickreduce/quick_reduce.h", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/quickreduce/quick_reduce.h#L208-L230" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/device_communicators/quick_all_reduce.py", "first": 49, "last": 62, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/quick_all_reduce.py#L49-L62" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce.py", "first": 260, "last": 283, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/custom_all_reduce.py#L260-L283" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "first": 310, "last": 324, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py#L310-L324" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py", "first": 216, "last": 220, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py#L216-L220" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/layers/k3_ar_fusion.py", "first": 36, "last": 38, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/k3_ar_fusion.py#L36-L38" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/distributed/device_communicators/pymscclpp.py", "first": 328, "last": 353, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/pymscclpp.py#L328-L353" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/device_communicators/pynccl.py", "first": 186, "last": 196, "engine": "vllm", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/pynccl.py#L186-L196" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/parallel_state.py", "first": 653, "last": 660, "engine": "vllm", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/parallel_state.py#L653-L660" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/layers/flashinfer_comm_fusion.py", "first": 852, "last": 864, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/flashinfer_comm_fusion.py#L852-L864" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/layers/communicator.py", "first": 809, "last": 866, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/communicator.py#L809-L866" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/layers/communicator.py", "first": 162, "last": 179, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/communicator.py#L162-L179" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/kernels/ops/kimi_k3/gemm_ar.py", "first": 1, "last": 15, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/kimi_k3/gemm_ar.py#L1-L15" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/layers/k3_gemm_ar.py", "first": 36, "last": 44, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/k3_gemm_ar.py#L36-L44" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/layers/k3_gemm_ar.py", "first": 50, "last": 55, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/k3_gemm_ar.py#L50-L55" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/kernels/ops/kimi_k3/gemm_ar.py", "first": 56, "last": 60, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/kimi_k3/gemm_ar.py#L56-L60" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_all_gather_reduce_scatter.cuh", "first": 13, "last": 31, "engine": null, "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/custom_all_gather_reduce_scatter.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/custom_all_gather_reduce_scatter.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_gather_reduce_scatter.cuh#L13-L31" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/utils/watchdog.py", "first": 152, "last": 163, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/watchdog.py#L152-L163" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/layers/flashinfer_comm_fusion.py", "first": 888, "last": 893, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/flashinfer_comm_fusion.py#L888-L893" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "python/sglang/srt/server_args.py", "first": 8351, "last": 8364, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L8351-L8364" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/device_communicators/all_reduce_utils.py", "first": 236, "last": 263, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/all_reduce_utils.py#L236-L263" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "vllm/distributed/device_communicators/all_reduce_utils.py", "first": 404, "last": 408, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/all_reduce_utils.py#L404-L408" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "docs/usage/troubleshooting.md", "first": 155, "last": 155, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/usage/troubleshooting.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/usage/troubleshooting.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/usage/troubleshooting.md#L155-L155" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_all_reduce.cuh", "first": 270, "last": 285, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/custom_all_reduce.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/custom_all_reduce.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_reduce.cuh#L270-L285" }, { "page": "05-parallelism/05-04-collectives-and-topology.html", "path": "csrc/custom_all_reduce.cuh", "first": 59, "last": 63, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/custom_all_reduce.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/custom_all_reduce.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_reduce.cuh#L59-L63" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 426, "last": 441, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L426-L441" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/envs.py", "first": 243, "last": 243, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L243-L243" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/utils/watchdog.py", "first": 149, "last": 163, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/watchdog.py#L149-L163" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/envs.py", "first": 243, "last": 243, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L243-L243" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/server_args.py", "first": 1261, "last": 1265, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1261-L1265" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/envs.py", "first": 242, "last": 242, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L242-L242" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/envs.py", "first": 1771, "last": 1775, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1771-L1775" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/layers/sampler.py", "first": 497, "last": 513, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L497-L513" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/abstract.py", "first": 60, "last": 81, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L60-L81" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 151, "last": 169, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L151-L169" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/distributed/device_communicators/shm_broadcast.py", "first": 488, "last": 533, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/shm_broadcast.py#L488-L533" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 715, "last": 719, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L715-L719" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 1029, "last": 1045, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L1029-L1045" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "multiproc_executor.py", "first": 420, "last": 420, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/executor/multiproc_executor.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1212, "page": "05-parallelism/05-05-worker-orchestration.html", "section": "worked-trace", "engine": "vllm", "path": "vllm/v1/executor/multiproc_executor.py", "first": 416, "last": 424 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L420-L420" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "multiproc_executor.py", "first": 541, "last": 555, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/executor/multiproc_executor.py", "reviewed_symbol": "_get_output_rank", "resolved_engine": "vllm", "resolved_path": "vllm/v1/executor/multiproc_executor.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/executor/multiproc_executor.py", "symbol": "_get_output_rank", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L541-L555" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/ray_executor_v2.py", "first": 219, "last": 231, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/ray_executor_v2.py#L219-L231" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/ray_executor.py", "first": 574, "last": 590, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/ray_executor.py#L574-L590" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 892, "last": 919, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L892-L919" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 5159, "last": 5176, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L5159-L5176" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1748, "last": 1771, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1748-L1771" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/managers/scheduler_components/request_receiver.py", "first": 209, "last": 216, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/request_receiver.py#L209-L216" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/managers/scheduler_components/ipc_channels.py", "first": 36, "last": 40, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/ipc_channels.py#L36-L40" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/engine/core.py", "first": 583, "last": 595, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L583-L595" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "multiproc_executor.py", "first": 340, "last": 351, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/executor/multiproc_executor.py", "reviewed_symbol": "collective_rpc", "resolved_engine": "vllm", "resolved_path": "vllm/v1/executor/multiproc_executor.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/executor/multiproc_executor.py", "symbol": "collective_rpc", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L340-L351" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 416, "last": 424, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L416-L424" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 383, "last": 393, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L383-L393" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1423, "last": 1437, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1423-L1437" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/distributed/parallel_state.py", "first": 2023, "last": 2027, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/parallel_state.py#L2023-L2027" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 2275, "last": 2288, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L2275-L2288" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/utils/common.py", "first": 539, "last": 546, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L539-L546" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/uniproc_executor.py", "first": 199, "last": 207, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/uniproc_executor.py#L199-L207" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/server_args.py", "first": 1015, "last": 1024, "engine": "sglang", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1015-L1024" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/distributed/bootstrap.py", "first": 169, "last": 184, "engine": "sglang", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/bootstrap.py#L169-L184" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 1824, "last": 1856, "engine": "sglang", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1824-L1856" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/server_args.py", "first": 9225, "last": 9228, "engine": "sglang", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9225-L9228" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/engine/arg_utils.py", "first": 1029, "last": 1032, "engine": "vllm", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L1029-L1032" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/entrypoints/cli/serve.py", "first": 208, "last": 223, "engine": "vllm", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/cli/serve.py#L208-L223" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 620, "last": 629, "engine": "vllm", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L620-L629" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/engine/arg_utils.py", "first": 2109, "last": 2123, "engine": "vllm", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2109-L2123" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/config/parallel.py", "first": 953, "last": 958, "engine": "vllm", "section": "multi-node", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L953-L958" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 305, "last": 324, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L305-L324" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 5178, "last": 5188, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L5178-L5188" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/utils/watchdog.py", "first": 166, "last": 175, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/watchdog.py#L166-L175" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/fault_tolerance/engine_core_sentinel.py", "first": 173, "last": 195, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/fault_tolerance/engine_core_sentinel.py#L173-L195" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/fault_tolerance/engine_core_sentinel.py", "first": 92, "last": 98, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/fault_tolerance/engine_core_sentinel.py#L92-L98" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "engine_core_sentinel.py", "first": 139, "last": 171, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/fault_tolerance/engine_core_sentinel.py", "reviewed_symbol": "_reinit_dp_group", "resolved_engine": "vllm", "resolved_path": "vllm/v1/fault_tolerance/engine_core_sentinel.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/fault_tolerance/engine_core_sentinel.py", "symbol": "_reinit_dp_group", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/fault_tolerance/engine_core_sentinel.py#L139-L171" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/v1/worker/sentinel/gpu_worker_sentinel.py", "first": 52, "last": 69, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/sentinel/gpu_worker_sentinel.py#L52-L69" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/utils/rank_consensus_checker.py", "first": 412, "last": 430, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/rank_consensus_checker.py#L412-L430" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/environ.py", "first": 332, "last": 332, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L332-L332" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 495, "last": 498, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L495-L498" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 1753, "last": 1753, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L1753-L1753" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/platforms/cuda.py", "first": 214, "last": 214, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L214-L214" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/platforms/xpu.py", "first": 111, "last": 111, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/xpu.py#L111-L111" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "vllm/platforms/cpu.py", "first": 48, "last": 48, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cpu.py#L48-L48" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "python/sglang/srt/distributed/parallel_state.py", "first": 2157, "last": 2164, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/parallel_state.py#L2157-L2164" }, { "page": "05-parallelism/05-05-worker-orchestration.html", "path": "multiproc_executor.py", "first": 1075, "last": 1084, "engine": null, "section": "hands-on", "declared_engine": "vllm", "declared_path": "vllm/v1/executor/multiproc_executor.py", "reviewed_symbol": "process_name", "resolved_engine": "vllm", "resolved_path": "vllm/v1/executor/multiproc_executor.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/executor/multiproc_executor.py", "symbol": "process_name", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L1075-L1084" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/model_executor/layers/utils.py", "first": 66, "last": 83, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/utils.py#L66-L83" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/penalties.py", "first": 27, "last": 28, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/penalties.py#L27-L28" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/config/vllm.py", "first": 648, "last": 700, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L700" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/config/vllm.py", "first": 725, "last": 743, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L725-L743" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/config/vllm.py", "first": 2438, "last": 2543, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L2438-L2543" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu/sample/min_p.py", "first": 34, "last": 34, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/sample/min_p.py#L34-L34" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_sampler.py", "first": 450, "last": 471, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_sampler.py#L450-L471" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/layers/sampler.py", "first": 597, "last": 598, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L597-L598" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/sampler.py", "first": 21, "last": 59, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L21-L59" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/layers/sampler.py", "first": 581, "last": 587, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L581-L587" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 400, "last": 417, "engine": "vllm", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L400-L417" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 1126, "last": 1132, "engine": "vllm", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L1126-L1132" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 860, "last": 871, "engine": "vllm", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L860-L871" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/metadata.py", "first": 20, "last": 21, "engine": "vllm", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/metadata.py#L20-L21" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu/sample/penalties.py", "first": 123, "last": 136, "engine": "vllm", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/sample/penalties.py#L123-L136" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu/sample/gumbel.py", "first": 29, "last": 31, "engine": "vllm", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/sample/gumbel.py#L29-L31" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/sampling/sampling_batch_info.py", "first": 203, "last": 208, "engine": "sglang", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_batch_info.py#L203-L208" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/logits_processor/interface.py", "first": 36, "last": 57, "engine": "vllm", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/interface.py#L36-L57" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/sampling/sampling_batch_info.py", "first": 307, "last": 317, "engine": "sglang", "section": "heterogeneous-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_batch_info.py#L307-L317" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_sampler.py", "first": 374, "last": 378, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_sampler.py#L374-L378" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_sampler.py", "first": 411, "last": 430, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_sampler.py#L411-L430" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_triton.py", "first": 127, "last": 200, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_triton.py#L127-L200" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_triton.py", "first": 933, "last": 947, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_triton.py#L933-L947" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_sampler.py", "first": 355, "last": 364, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_sampler.py#L355-L364" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "csrc/libtorch_stable/sampler.cu", "first": 549, "last": 569, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/sampler.cu#L549-L569" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "csrc/libtorch_stable/sampler.cu", "first": 573, "last": 596, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/sampler.cu#L573-L596" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "csrc/libtorch_stable/sampler.cu", "first": 717, "last": 751, "engine": null, "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/sampler.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/sampler.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/sampler.cu#L717-L751" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "csrc/libtorch_stable/sampler.cu", "first": 846, "last": 866, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/sampler.cu#L846-L866" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "csrc/libtorch_stable/torch_bindings.cpp", "first": 556, "last": 565, "engine": null, "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/torch_bindings.cpp", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/torch_bindings.cpp", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/torch_bindings.cpp#L556-L565" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/_custom_ops.py", "first": 3055, "last": 3096, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/_custom_ops.py#L3055-L3096" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/model_executor/layers/sparse_attn_indexer.py", "first": 509, "last": 512, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/sparse_attn_indexer.py#L509-L512" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/attention/ops/rocm_aiter_mla_sparse.py", "first": 969, "last": 985, "engine": "vllm", "section": "selection", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/rocm_aiter_mla_sparse.py#L969-L985" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/sampling_params.py", "first": 745, "last": 750, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L745-L750" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/sampling_params.py", "first": 529, "last": 534, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L529-L534" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/sampler.py", "first": 235, "last": 238, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L235-L238" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/sampler.py", "first": 298, "last": 303, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L298-L303" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/sampling/sampling_params.py", "first": 143, "last": 149, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_params.py#L143-L149" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/layers/sampler.py", "first": 143, "last": 145, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L143-L145" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/metadata.py", "first": 23, "last": 23, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/metadata.py#L23-L23" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 1331, "last": 1338, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L1331-L1338" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_sampler.py", "first": 164, "last": 170, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_sampler.py#L164-L170" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/layers/sampler.py", "first": 709, "last": 729, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L709-L729" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu/sample/states.py", "first": 50, "last": 54, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/sample/states.py#L50-L54" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/worker/gpu/sample/gumbel.py", "first": 97, "last": 99, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/sample/gumbel.py#L97-L99" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2031, "last": 2038, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2031-L2038" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/sampler.py", "first": 86, "last": 94, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L86-L94" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/logits_processor/builtin.py", "first": 102, "last": 104, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/builtin.py#L102-L104" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/logprobs.py", "first": 25, "last": 27, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/logprobs.py#L25-L27" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/model_executor/model_runner.py", "first": 1745, "last": 1760, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner.py#L1745-L1760" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/layers/sampler.py", "first": 207, "last": 215, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L207-L215" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/entrypoints/generate/beam_search/offline.py", "first": 115, "last": 125, "engine": "vllm", "section": "beam-and-logprobs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/generate/beam_search/offline.py#L115-L125" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/sampling_params.py", "first": 1246, "last": 1261, "engine": "vllm", "section": "beam-and-logprobs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L1246-L1261" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/layers/logprob_processor.py", "first": 93, "last": 94, "engine": "sglang", "section": "beam-and-logprobs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/logprob_processor.py#L93-L94" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/sampler.py", "first": 335, "last": 335, "engine": "vllm", "section": "beam-and-logprobs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L335-L335" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/config/model.py", "first": 250, "last": 250, "engine": "vllm", "section": "beam-and-logprobs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L250-L250" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/utils/async_probe.py", "first": 66, "last": 76, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/async_probe.py#L66-L76" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/environ.py", "first": 1091, "last": 1091, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1091-L1091" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/sampling_params.py", "first": 992, "last": 996, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L992-L996" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "python/sglang/srt/sampling/penaltylib/frequency_penalty.py", "first": 18, "last": 22, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/penaltylib/frequency_penalty.py#L18-L22" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "vllm/v1/sample/ops/topk_topp_triton.py", "first": 134, "last": 136, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_triton.py#L134-L136" }, { "page": "06-decoding-algorithms/06-01-sampling-on-gpu.html", "path": "topk_topp_sampler.py", "first": 106, "last": 107, "engine": null, "section": "further-reading", "declared_engine": "vllm", "declared_path": "vllm/v1/sample/ops/topk_topp_sampler.py", "reviewed_symbol": "PowerPC", "resolved_engine": "vllm", "resolved_path": "vllm/v1/sample/ops/topk_topp_sampler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/sample/ops/topk_topp_sampler.py", "symbol": "PowerPC", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/ops/topk_topp_sampler.py#L106-L107" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 674, "last": 683, "engine": "sglang", "section": "the-algorithm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L674-L683" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 394, "last": 507, "engine": "vllm", "section": "the-algorithm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L394-L507" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 743, "last": 761, "engine": "vllm", "section": "greedy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L743-L761" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 859, "last": 872, "engine": "sglang", "section": "greedy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L859-L872" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/spec_decode/utils.py", "first": 598, "last": 601, "engine": "vllm", "section": "economics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/utils.py#L598-L601" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/config/speculative.py", "first": 180, "last": 186, "engine": "vllm", "section": "economics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L180-L186" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1265, "last": 1271, "engine": "vllm", "section": "economics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1265-L1271" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/config/vllm.py", "first": 2465, "last": 2467, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L2465-L2467" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/spec_decode/metadata.py", "first": 9, "last": 27, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/metadata.py#L9-L27" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 806, "last": 838, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L806-L838" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/config/speculative.py", "first": 289, "last": 296, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L289-L296" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 919, "last": 932, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L919-L932" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 770, "last": 776, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L770-L776" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/server_args.py", "first": 2192, "last": 2196, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2192-L2196" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/kernels/ops/speculative/reject_sampling.py", "first": 52, "last": 87, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/speculative/reject_sampling.py#L52-L87" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/kernels/ops/speculative/reject_sampling.py", "first": 90, "last": 119, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/speculative/reject_sampling.py#L90-L119" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2928, "last": 2936, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2928-L2936" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1847, "last": 1859, "engine": "vllm", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1847-L1859" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 556, "last": 565, "engine": "vllm", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L556-L565" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 504, "last": 510, "engine": "vllm", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L504-L510" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 505, "last": 506, "engine": "vllm", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L505-L506" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/mem_cache/allocation_sizing.py", "first": 27, "last": 54, "engine": "sglang", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation_sizing.py#L27-L54" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 694, "last": 708, "engine": "sglang", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L694-L708" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 2801, "last": 2819, "engine": "sglang", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L2801-L2819" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/speculative/eagle_worker_common.py", "first": 406, "last": 434, "engine": "sglang", "section": "kv-rollback", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_common.py#L406-L434" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 639, "last": 648, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L639-L648" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 610, "last": 618, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L610-L618" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/server_args.py", "first": 2182, "last": 2196, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2182-L2196" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 810, "last": 816, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L810-L816" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "python/sglang/kernels/aot/csrc/speculative/speculative_sampling.cuh", "first": 77, "last": 94, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/aot/csrc/speculative/speculative_sampling.cuh#L77-L94" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 38, "last": 59, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L38-L59" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 510, "last": 564, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L510-L564" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 563, "last": 565, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L563-L565" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2313, "last": 2317, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2313-L2317" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 650, "last": 653, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L650-L653" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/spec_decode/metrics.py", "first": 113, "last": 118, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/metrics.py#L113-L118" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 873, "last": 953, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L873-L953" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1841, "last": 1859, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1841-L1859" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "kv_cache_manager.py", "first": 556, "last": 565, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/kv_cache_manager.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1323, "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "section": "kv-rollback", "engine": "vllm", "path": "vllm/v1/core/kv_cache_manager.py", "first": 556, "last": 565 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L556-L565" }, { "page": "06-decoding-algorithms/06-02-speculative-decoding.html", "path": "vllm/config/speculative.py", "first": 219, "last": 226, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L219-L226" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/config/speculative.py", "first": 69, "last": 80, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L69-L80" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/draft_model.py", "first": 63, "last": 78, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/draft_model.py#L63-L78" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 1745, "last": 1752, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L1745-L1752" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/config/speculative.py", "first": 1418, "last": 1433, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L1418-L1433" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/vocab_mapping.py", "first": 103, "last": 120, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/vocab_mapping.py#L103-L120" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "docs/features/speculative_decoding/draft_model.md", "first": 79, "last": 84, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/speculative_decoding/draft_model.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/speculative_decoding/draft_model.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/speculative_decoding/draft_model.md#L79-L84" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/ngram_proposer.py", "first": 214, "last": 241, "engine": "vllm", "section": "ngram", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/ngram_proposer.py#L214-L241" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/ngram_proposer.py", "first": 255, "last": 263, "engine": "vllm", "section": "ngram", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/ngram_proposer.py#L255-L263" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/ngram_proposer_gpu.py", "first": 56, "last": 58, "engine": "vllm", "section": "ngram", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/ngram_proposer_gpu.py#L56-L58" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/config/speculative.py", "first": 815, "last": 820, "engine": "vllm", "section": "ngram", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L815-L820" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py", "first": 27, "last": 32, "engine": null, "section": "ngram", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py#L27-L32" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "examples/features/speculative_decoding/spec_decode_offline.py", "first": 55, "last": 56, "engine": null, "section": "ngram", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "examples/features/speculative_decoding/spec_decode_offline.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "examples/features/speculative_decoding/spec_decode_offline.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/features/speculative_decoding/spec_decode_offline.py#L55-L56" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 7936, "last": 7949, "engine": "vllm", "section": "ngram-gpu", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L7936-L7949" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/ngram_proposer.py", "first": 39, "last": 53, "engine": "vllm", "section": "ngram-gpu", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/ngram_proposer.py#L39-L53" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/ngram_proposer_gpu.py", "first": 84, "last": 108, "engine": "vllm", "section": "ngram-gpu", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/ngram_proposer_gpu.py#L84-L108" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 4299, "last": 4314, "engine": "vllm", "section": "ngram-gpu", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4299-L4314" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/suffix_decoding.py", "first": 77, "last": 89, "engine": "vllm", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/suffix_decoding.py#L77-L89" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/config/speculative.py", "first": 205, "last": 209, "engine": "vllm", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L205-L209" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "docs/features/speculative_decoding/suffix.md", "first": 12, "last": 13, "engine": null, "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/speculative_decoding/suffix.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/speculative_decoding/suffix.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/speculative_decoding/suffix.md#L12-L13" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.h", "first": 20, "last": 28, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.h#L20-L28" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "suffix_automaton.cpp", "first": 98, "last": 112, "engine": null, "section": "suffix", "declared_engine": "sglang", "declared_path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "reviewed_symbol": "SuffixAutomaton", "resolved_engine": "sglang", "resolved_path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "symbol": "SuffixAutomaton", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp#L98-L112" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "first": 146, "last": 178, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp#L146-L178" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "first": 183, "last": 193, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp#L183-L193" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/server_args.py", "first": 2327, "last": 2342, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2327-L2342" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/speculative/ngram_info.py", "first": 11, "last": 41, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/ngram_info.py#L11-L41" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/speculative/ngram_worker.py", "first": 416, "last": 423, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/ngram_worker.py#L416-L423" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/server_args.py", "first": 2346, "last": 2359, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2346-L2359" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 729, "last": 747, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L729-L747" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/speculative/external_corpus_manager.py", "first": 40, "last": 52, "engine": "sglang", "section": "suffix", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/external_corpus_manager.py#L40-L52" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/speculative/ngram_worker.py", "first": 238, "last": 238, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/ngram_worker.py#L238-L238" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 5138, "last": 5149, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L5138-L5149" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/server_args.py", "first": 9569, "last": 9583, "engine": "sglang", "section": "decision", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9569-L9583" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py", "first": 33, "last": 36, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/e2e/spec_decode/ngram_suffix/test_ngram_suffix.py#L33-L36" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/metrics.py", "first": 113, "last": 118, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/metrics.py#L113-L118" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/vocab_mapping.py", "first": 122, "last": 136, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/vocab_mapping.py#L122-L136" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/ngram_proposer.py", "first": 283, "last": 293, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/ngram_proposer.py#L283-L293" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/ngram_proposer_gpu.py", "first": 96, "last": 102, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/ngram_proposer_gpu.py#L96-L102" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "suffix_automaton.cpp", "first": 181, "last": 215, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "reviewed_symbol": "Result", "resolved_engine": "sglang", "resolved_path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp", "symbol": "Result", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/ngram_corpus/suffix_automaton.cpp#L181-L215" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 717, "last": 723, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L717-L723" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "vllm/v1/spec_decode/suffix_decoding.py", "first": 10, "last": 14, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/suffix_decoding.py#L10-L14" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "docs/features/speculative_decoding/n_gram.md", "first": 3, "last": 4, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/features/speculative_decoding/n_gram.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/features/speculative_decoding/n_gram.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/speculative_decoding/n_gram.md#L3-L4" }, { "page": "06-decoding-algorithms/06-03-draft-sources.html", "path": "gpu_model_runner.py", "first": 4299, "last": 4315, "engine": null, "section": "further-reading", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "use_ngram_gpu", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "use_ngram_gpu", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4299-L4315" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/spec_decode/medusa.py", "first": 48, "last": 58, "engine": "vllm", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/medusa.py#L48-L58" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/medusa.py", "first": 105, "last": 106, "engine": "vllm", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/medusa.py#L105-L106" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/medusa.py", "first": 45, "last": 46, "engine": "vllm", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/medusa.py#L45-L46" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/config/speculative.py", "first": 69, "last": 79, "engine": "vllm", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L69-L79" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 693, "last": 694, "engine": "vllm", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L693-L694" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/config/speculative.py", "first": 895, "last": 901, "engine": "vllm", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L895-L901" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/medusa.py", "first": 70, "last": 90, "engine": "vllm", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/medusa.py#L70-L90" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/server_args.py", "first": 2101, "last": 2105, "engine": "sglang", "section": "medusa", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2101-L2105" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/llama_eagle.py", "first": 95, "last": 124, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama_eagle.py#L95-L124" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/llama_eagle.py", "first": 40, "last": 44, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama_eagle.py#L40-L44" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/spec_decode/eagle.py", "first": 10, "last": 22, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/eagle.py#L10-L22" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 688, "last": 694, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L688-L694" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 769, "last": 772, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L769-L772" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/_custom_ops.py", "first": 3999, "last": 4020, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/_custom_ops.py#L3999-L4020" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/attention/backends/mla/amx_mla.py", "first": 377, "last": 397, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/amx_mla.py#L377-L397" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 1645, "last": 1646, "engine": "vllm", "section": "eagle", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L1645-L1646" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 306, "last": 325, "engine": "sglang", "section": "eagle23", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L306-L325" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 107, "last": 126, "engine": "sglang", "section": "eagle23", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L107-L126" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 821, "last": 822, "engine": "sglang", "section": "eagle23", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L821-L822" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "speculative_hook.py", "first": 606, "last": 619, "engine": null, "section": "eagle23", "declared_engine": "sglang", "declared_path": "python/sglang/srt/arg_groups/speculative_hook.py", "reviewed_symbol": "server_args", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/arg_groups/speculative_hook.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "symbol": "server_args", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L606-L619" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/llama_eagle3.py", "first": 183, "last": 187, "engine": "vllm", "section": "eagle23", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama_eagle3.py#L183-L187" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/llama_eagle3.py", "first": 359, "last": 379, "engine": "vllm", "section": "eagle23", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama_eagle3.py#L359-L379" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "llama_eagle3.py", "first": 176, "last": 182, "engine": null, "section": "eagle23", "declared_engine": "vllm", "declared_path": "vllm/model_executor/models/llama_eagle3.py", "reviewed_symbol": "num_aux_hidden_states", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/models/llama_eagle3.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/models/llama_eagle3.py", "symbol": "num_aux_hidden_states", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama_eagle3.py#L176-L182" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 548, "last": 550, "engine": "vllm", "section": "eagle23", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L548-L550" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/llama_eagle3.py", "first": 294, "last": 307, "engine": "vllm", "section": "eagle23", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama_eagle3.py#L294-L307" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "llama_eagle3.py", "first": 387, "last": 390, "engine": null, "section": "eagle23", "declared_engine": "vllm", "declared_path": "vllm/model_executor/models/llama_eagle3.py", "reviewed_symbol": "continue", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/models/llama_eagle3.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/models/llama_eagle3.py", "symbol": "continue", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama_eagle3.py#L387-L390" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/model_executor/models/deepseek_mtp.py", "first": 106, "last": 137, "engine": "vllm", "section": "mtp", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_mtp.py#L106-L137" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "deepseek_mtp.py", "first": 209, "last": 212, "engine": null, "section": "mtp", "declared_engine": "vllm", "declared_path": "vllm/model_executor/models/deepseek_mtp.py", "reviewed_symbol": "inputs_embeds", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/models/deepseek_mtp.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/models/deepseek_mtp.py", "symbol": "inputs_embeds", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_mtp.py#L209-L212" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "vllm/config/speculative.py", "first": 37, "last": 56, "engine": "vllm", "section": "mtp", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L37-L56" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 819, "last": 843, "engine": "sglang", "section": "mtp", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L819-L843" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 226, "last": 285, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L226-L285" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/kernels/aot/csrc/speculative/eagle_utils.cu", "first": 34, "last": 123, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/aot/csrc/speculative/eagle_utils.cu#L34-L123" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/kernels/ops/speculative/spec_tree.py", "first": 108, "last": 134, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/speculative/spec_tree.py#L108-L134" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "spec_tree.py", "first": 170, "last": 173, "engine": null, "section": "tree-attention", "declared_engine": "sglang", "declared_path": "python/sglang/kernels/ops/speculative/spec_tree.py", "reviewed_symbol": "positions_ptr", "resolved_engine": "sglang", "resolved_path": "python/sglang/kernels/ops/speculative/spec_tree.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/kernels/ops/speculative/spec_tree.py", "symbol": "positions_ptr", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/speculative/spec_tree.py#L170-L173" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/kernels/ops/speculative/spec_tree.py", "first": 136, "last": 168, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/speculative/spec_tree.py#L136-L168" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 296, "last": 300, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L296-L300" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/kernels/ops/speculative/spec_tree.py", "first": 203, "last": 237, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/speculative/spec_tree.py#L203-L237" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 772, "last": 775, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L772-L775" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_info.py", "first": 43, "last": 48, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_info.py#L43-L48" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 527, "last": 534, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L527-L534" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_worker_common.py", "first": 406, "last": 435, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_common.py#L406-L435" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 723, "last": 756, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L723-L756" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "first": 2197, "last": 2208, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L2197-L2208" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_info.py", "first": 82, "last": 136, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_info.py#L82-L136" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 179, "last": 184, "engine": "sglang", "section": "tree-attention", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L179-L184" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_utils.py", "first": 193, "last": 199, "engine": null, "section": "tree-attention", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_utils.py", "reviewed_symbol": "tree_mask_mode", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_utils.py", "symbol": "tree_mask_mode", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L193-L199" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_utils.py", "first": 145, "last": 149, "engine": null, "section": "tree-attention", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_utils.py", "reviewed_symbol": "default_tree_mask_mode", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_utils.py", "symbol": "default_tree_mask_mode", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L145-L149" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py", "first": 440, "last": 462, "engine": "sglang", "section": "cost-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py#L440-L462" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/draft_utils.py", "first": 71, "last": 78, "engine": "sglang", "section": "cost-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/draft_utils.py#L71-L78" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_worker_v2.py", "first": 501, "last": 512, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_worker_v2.py", "reviewed_symbol": "draft", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_worker_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_worker_v2.py", "symbol": "draft", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_v2.py#L501-L512" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_worker_v2.py", "first": 621, "last": 633, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_worker_v2.py", "reviewed_symbol": "speculative_num_steps", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_worker_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_worker_v2.py", "symbol": "speculative_num_steps", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_v2.py#L621-L633" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_utils.py", "first": 64, "last": 86, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_utils.py", "reviewed_symbol": "per_step_draft_out_cache_loc", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_utils.py", "symbol": "per_step_draft_out_cache_loc", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L64-L86" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_utils.py", "first": 727, "last": 742, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_utils.py", "reviewed_symbol": "Sample", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_utils.py", "symbol": "Sample", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L727-L742" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_worker_common.py", "first": 612, "last": 622, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_common.py#L612-L622" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 711, "last": 713, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L711-L713" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 76, "last": 80, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L76-L80" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 621, "last": 625, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L621-L625" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 685, "last": 699, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L685-L699" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/draft_utils.py", "first": 98, "last": 104, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/draft_utils.py#L98-L104" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 821, "last": 825, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L821-L825" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_worker_common.py", "first": 350, "last": 356, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_worker_common.py", "reviewed_symbol": "verify_mask", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_worker_common.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_worker_common.py", "symbol": "verify_mask", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_common.py#L350-L356" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "python/sglang/srt/environ.py", "first": 468, "last": 468, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L468-L468" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "speculative_hook.py", "first": 819, "last": 843, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/arg_groups/speculative_hook.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1416, "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "section": "mtp", "engine": "sglang", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 819, "last": 843 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L819-L843" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_worker_common.py", "first": 438, "last": 458, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_worker_common.py", "reviewed_symbol": "_compact_accept_to_front", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_worker_common.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_worker_common.py", "symbol": "_compact_accept_to_front", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_common.py#L438-L458" }, { "page": "06-decoding-algorithms/06-04-eagle-medusa-mtp.html", "path": "eagle_worker_common.py", "first": 450, "last": 450, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_worker_common.py", "reviewed_symbol": "trailing", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_worker_common.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_worker_common.py", "symbol": "trailing", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_common.py#L450-L450" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_xgrammar.py", "first": 350, "last": 353, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L350-L353" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/envs.py", "first": 1642, "last": 1643, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1642-L1643" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/environ.py", "first": 374, "last": 375, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L374-L375" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_xgrammar.py", "first": 66, "last": 71, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L66-L71" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/envs.py", "first": 1637, "last": 1637, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1637-L1637" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/kernels/ops/grammar/bitmask_ops.py", "first": 70, "last": 83, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/grammar/bitmask_ops.py#L70-L83" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/xgrammar_backend.py", "first": 61, "last": 70, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/xgrammar_backend.py#L61-L70" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/engine/core.py", "first": 594, "last": 603, "engine": "vllm", "section": "where-the-cost-lands", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L594-L603" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/__init__.py", "first": 61, "last": 69, "engine": "vllm", "section": "where-the-cost-lands", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L61-L69" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/sampling/sampling_batch_info.py", "first": 239, "last": 264, "engine": "sglang", "section": "where-the-cost-lands", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_batch_info.py#L239-L264" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 88, "last": 94, "engine": "sglang", "section": "where-the-cost-lands", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L88-L94" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/llguidance_backend.py", "first": 151, "last": 159, "engine": "sglang", "section": "where-the-cost-lands", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/llguidance_backend.py#L151-L159" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1879, "last": 1903, "engine": "sglang", "section": "where-the-cost-lands", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1879-L1903" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 2247, "last": 2251, "engine": "sglang", "section": "where-the-cost-lands", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L2247-L2251" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_types.py", "first": 31, "last": 142, "engine": "vllm", "section": "backend-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_types.py#L31-L142" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 241, "last": 282, "engine": "sglang", "section": "backend-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L241-L282" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_xgrammar.py", "first": 79, "last": 131, "engine": null, "section": "backend-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_xgrammar.py", "reviewed_symbol": "compile_grammar", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_xgrammar.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_xgrammar.py", "symbol": "compile_grammar", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L79-L131" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_xgrammar.py", "first": 329, "last": 339, "engine": null, "section": "backend-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_xgrammar.py", "reviewed_symbol": "choice_as_grammar", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_xgrammar.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_xgrammar.py", "symbol": "choice_as_grammar", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L329-L339" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_outlines.py", "first": 209, "last": 213, "engine": "vllm", "section": "backend-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_outlines.py#L209-L213" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_lm_format_enforcer.py", "first": 191, "last": 195, "engine": null, "section": "backend-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_lm_format_enforcer.py", "reviewed_symbol": "so_params", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_lm_format_enforcer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_lm_format_enforcer.py", "symbol": "so_params", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_lm_format_enforcer.py#L191-L195" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/outlines_backend.py", "first": 160, "last": 164, "engine": "sglang", "section": "backend-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/outlines_backend.py#L160-L164" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/server_args.py", "first": 6222, "last": 6224, "engine": "sglang", "section": "backend-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L6222-L6224" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "server_args.py", "first": 245, "last": 245, "engine": null, "section": "backend-inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "GRAMMAR_BACKEND_CHOICES", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "GRAMMAR_BACKEND_CHOICES", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L245-L245" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/config/structured_outputs.py", "first": 21, "last": 21, "engine": "vllm", "section": "backend-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/structured_outputs.py#L21-L21" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/sampling_params.py", "first": 1156, "last": 1193, "engine": "vllm", "section": "backend-inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L1156-L1193" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_types.py", "first": 19, "last": 25, "engine": null, "section": "backend-inventory", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_types.py", "reviewed_symbol": "StructuredOutputOptions", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_types.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_types.py", "symbol": "StructuredOutputOptions", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_types.py#L19-L25" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 120, "last": 146, "engine": "sglang", "section": "jump-forward", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L120-L146" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/outlines_jump_forward.py", "first": 98, "last": 107, "engine": "sglang", "section": "jump-forward", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/outlines_jump_forward.py#L98-L107" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/xgrammar_backend.py", "first": 174, "last": 196, "engine": "sglang", "section": "jump-forward", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/xgrammar_backend.py#L174-L196" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/outlines_backend.py", "first": 157, "last": 158, "engine": "sglang", "section": "jump-forward", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/outlines_backend.py#L157-L158" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_guidance.py", "first": 177, "last": 182, "engine": null, "section": "jump-forward", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_guidance.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1521, "page": "06-decoding-algorithms/06-05-structured-decoding.html", "section": "further-reading", "engine": "vllm", "path": "vllm/v1/structured_output/backend_guidance.py", "first": 177, "last": 182 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_guidance.py#L177-L182" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_xgrammar.py", "first": 146, "last": 146, "engine": null, "section": "jump-forward", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_xgrammar.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1520, "page": "06-decoding-algorithms/06-05-structured-decoding.html", "section": "further-reading", "engine": "vllm", "path": "vllm/v1/structured_output/backend_xgrammar.py", "first": 146, "last": 146 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L146-L146" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/__init__.py", "first": 233, "last": 242, "engine": "vllm", "section": "spec-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L233-L242" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/__init__.py", "first": 298, "last": 359, "engine": "vllm", "section": "spec-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L298-L359" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2309, "last": 2317, "engine": "vllm", "section": "spec-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2309-L2317" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_xgrammar.py", "first": 181, "last": 201, "engine": null, "section": "spec-decode", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_xgrammar.py", "reviewed_symbol": "validate_tokens", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_xgrammar.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_xgrammar.py", "symbol": "validate_tokens", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L181-L201" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1867, "last": 1876, "engine": "vllm", "section": "spec-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1867-L1876" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_xgrammar.py", "first": 124, "last": 128, "engine": null, "section": "spec-decode", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_xgrammar.py", "reviewed_symbol": "max_rollback_tokens", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_xgrammar.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_xgrammar.py", "symbol": "max_rollback_tokens", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L124-L128" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_lm_format_enforcer.py", "first": 134, "last": 137, "engine": null, "section": "spec-decode", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_lm_format_enforcer.py", "reviewed_symbol": "max_rollback_tokens", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_lm_format_enforcer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_lm_format_enforcer.py", "symbol": "max_rollback_tokens", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_lm_format_enforcer.py#L134-L137" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 467, "last": 495, "engine": "sglang", "section": "spec-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L467-L495" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/xgrammar_backend.py", "first": 58, "last": 58, "engine": "sglang", "section": "spec-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/xgrammar_backend.py#L58-L58" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 617, "last": 646, "engine": "sglang", "section": "spec-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L617-L646" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "spec_utils.py", "first": 549, "last": 554, "engine": null, "section": "spec-decode", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/spec_utils.py", "reviewed_symbol": "tree_traverse_time", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/spec_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/spec_utils.py", "symbol": "tree_traverse_time", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L549-L554" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/sampling_params.py", "first": 1156, "last": 1193, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L1156-L1193" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/request.py", "first": 114, "last": 115, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L114-L115" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/__init__.py", "first": 167, "last": 176, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L167-L176" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/request.py", "first": 50, "last": 59, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/request.py#L50-L59" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2817, "last": 2825, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2817-L2825" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "scheduler.py", "first": 1731, "last": 1733, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/scheduler.py", "reviewed_symbol": "is_prefill_chunk", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/scheduler.py", "symbol": "is_prefill_chunk", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1731-L1733" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "scheduler.py", "first": 1720, "last": 1742, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/scheduler.py", "reviewed_symbol": "get_grammar_bitmask", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/scheduler.py", "symbol": "get_grammar_bitmask", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1720-L1742" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_xgrammar.py", "first": 208, "last": 209, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_xgrammar.py", "reviewed_symbol": "fill_next_token_bitmask", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_xgrammar.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_xgrammar.py", "symbol": "fill_next_token_bitmask", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L208-L209" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "__init__.py", "first": 361, "last": 368, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/__init__.py", "reviewed_symbol": "bitmask_tensor", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/__init__.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/__init__.py", "symbol": "bitmask_tensor", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L361-L368" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/utils.py", "first": 114, "last": 162, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/utils.py#L114-L162" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "scheduler.py", "first": 1918, "last": 1943, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/scheduler.py", "reviewed_symbol": "should_advance", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/scheduler.py", "symbol": "should_advance", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1918-L1943" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/grammar_manager.py", "first": 131, "last": 182, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/grammar_manager.py#L131-L182" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "base_grammar_backend.py", "first": 284, "last": 293, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/constrained/base_grammar_backend.py", "reviewed_symbol": "get_cached_or_future_value", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/constrained/base_grammar_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "symbol": "get_cached_or_future_value", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L284-L293" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "grammar_manager.py", "first": 203, "last": 303, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/constrained/grammar_manager.py", "reviewed_symbol": "pp_rank", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/constrained/grammar_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/constrained/grammar_manager.py", "symbol": "pp_rank", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/grammar_manager.py#L203-L303" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/utils.py", "first": 71, "last": 81, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/utils.py#L71-L81" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/envs.py", "first": 1642, "last": 1643, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1642-L1643" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_outlines.py", "first": 335, "last": 343, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_outlines.py#L335-L343" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_xgrammar.py", "first": 265, "last": 278, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L265-L278" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_types.py", "first": 111, "last": 123, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_types.py#L111-L123" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/config/structured_outputs.py", "first": 26, "last": 30, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/structured_outputs.py#L26-L30" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/sampling_params.py", "first": 1034, "last": 1039, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L1034-L1039" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "benchmarks/run_structured_output_benchmark.sh", "first": 1, "last": 13, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "benchmarks/run_structured_output_benchmark.sh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "benchmarks/run_structured_output_benchmark.sh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/run_structured_output_benchmark.sh#L1-L13" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/sampling_params.py", "first": 1156, "last": 1193, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L1156-L1193" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "backend_guidance.py", "first": 50, "last": 73, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/structured_output/backend_guidance.py", "reviewed_symbol": "has_guidance_unsupported_json_features", "resolved_engine": "vllm", "resolved_path": "vllm/v1/structured_output/backend_guidance.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/structured_output/backend_guidance.py", "symbol": "has_guidance_unsupported_json_features", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_guidance.py#L50-L73" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "python/sglang/srt/constrained/outlines_jump_forward.py", "first": 16, "last": 16, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/outlines_jump_forward.py#L16-L16" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_xgrammar.py", "first": 146, "last": 146, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L146-L146" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_guidance.py", "first": 177, "last": 182, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_guidance.py#L177-L182" }, { "page": "06-decoding-algorithms/06-05-structured-decoding.html", "path": "vllm/v1/structured_output/backend_outlines.py", "first": 316, "last": 318, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_outlines.py#L316-L318" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "first": 22, "last": 47, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L22-L47" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "adaptive_runtime_state.py", "first": 103, "last": 107, "engine": null, "section": "the-problem", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/adaptive_runtime_state.py", "reviewed_symbol": "worker", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/adaptive_runtime_state.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/adaptive_runtime_state.py", "symbol": "worker", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_runtime_state.py#L103-L107" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 2278, "last": 2282, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2278-L2282" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "first": 50, "last": 86, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L50-L86" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 774, "last": 785, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L774-L785" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dspark_components/dspark_draft.py", "first": 328, "last": 334, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dspark_components/dspark_draft.py#L328-L334" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dflash_worker_v2.py", "first": 1814, "last": 1815, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_worker_v2.py#L1814-L1815" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 109, "last": 111, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L109-L111" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/spec_utils.py", "first": 343, "last": 355, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_utils.py#L343-L355" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/eagle_utils.py", "first": 151, "last": 214, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_utils.py#L151-L214" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 1645, "last": 1646, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L1645-L1646" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 147, "last": 151, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L147-L151" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 761, "last": 765, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L761-L765" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/__init__.py", "first": 51, "last": 58, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/__init__.py#L51-L58" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/__init__.py", "first": 8, "last": 58, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/__init__.py#L8-L58" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 634, "last": 704, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L634-L704" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/__init__.py", "first": 8, "last": 58, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/__init__.py#L8-L58" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/vllm.py", "first": 657, "last": 677, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L657-L677" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 1442, "last": 1453, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L1442-L1453" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 167, "last": 172, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L167-L172" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 1071, "last": 1072, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L1071-L1072" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/models/dflash.py", "first": 1, "last": 4, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/dflash.py#L1-L4" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dflash_worker_v2.py", "first": 405, "last": 410, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_worker_v2.py#L405-L410" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dflash_utils.py", "first": 782, "last": 785, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_utils.py#L782-L785" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dflash_utils.py", "first": 825, "last": 844, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_utils.py#L825-L844" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dflash_info.py", "first": 26, "last": 32, "engine": null, "section": "inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/dflash_info.py", "reviewed_symbol": "DFlashVerifyInput", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/dflash_info.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/dflash_info.py", "symbol": "DFlashVerifyInput", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_info.py#L26-L32" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dflash_info_v2.py", "first": 35, "last": 37, "engine": null, "section": "inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/dflash_info_v2.py", "reviewed_symbol": "DFlashDraftInputV2", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/dflash_info_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/dflash_info_v2.py", "symbol": "DFlashDraftInputV2", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_info_v2.py#L35-L37" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/models/dspark.py", "first": 59, "last": 71, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/dspark.py#L59-L71" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dspark_components/dspark_config.py", "first": 17, "last": 17, "engine": null, "section": "inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/dspark_components/dspark_config.py", "reviewed_symbol": "SUPPORTED_DSPARK_MARKOV_HEAD_TYPES", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/dspark_components/dspark_config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/dspark_components/dspark_config.py", "symbol": "SUPPORTED_DSPARK_MARKOV_HEAD_TYPES", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dspark_components/dspark_config.py#L17-L17" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dspark_config.py", "first": 16, "last": 16, "engine": null, "section": "inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/dspark_components/dspark_config.py", "reviewed_symbol": "DEFAULT_DSPARK_GAMMA", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/dspark_components/dspark_config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/dspark_components/dspark_config.py", "symbol": "DEFAULT_DSPARK_GAMMA", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dspark_components/dspark_config.py#L16-L16" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dspark_components/dspark_verify.py", "first": 697, "last": 760, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dspark_components/dspark_verify.py#L697-L760" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/spec_info.py", "first": 131, "last": 135, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_info.py#L131-L135" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/environ.py", "first": 1137, "last": 1137, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1137-L1137" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dspark_components/dspark_planner.py", "first": 132, "last": 172, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dspark_components/dspark_planner.py#L132-L172" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/ragged_verify.py", "first": 13, "last": 16, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/ragged_verify.py#L13-L16" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/models/deepseek_v4_dspark.py", "first": 91, "last": 91, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_v4_dspark.py#L91-L91" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/frozen_kv_mtp_utils.py", "first": 87, "last": 90, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/frozen_kv_mtp_utils.py#L87-L90" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/frozen_kv_mtp_worker_v2.py", "first": 648, "last": 654, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/frozen_kv_mtp_worker_v2.py#L648-L654" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 52, "last": 58, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L52-L58" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 2101, "last": 2105, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2101-L2105" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/spec_info.py", "first": 95, "last": 103, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_info.py#L95-L103" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/spec_decode/gemma4.py", "first": 45, "last": 48, "engine": "vllm", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/gemma4.py#L45-L48" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/multi_layer_eagle_worker_v2.py", "first": 188, "last": 191, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/multi_layer_eagle_worker_v2.py#L188-L191" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 719, "last": 724, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L719-L724" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 2270, "last": 2277, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2270-L2277" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "adaptive_spec_params.py", "first": 72, "last": 76, "engine": null, "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/adaptive_spec_params.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1526, "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "section": "the-problem", "engine": "sglang", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "first": 50, "last": 86 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L72-L76" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/standalone_worker_v2.py", "first": 35, "last": 36, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/standalone_worker_v2.py#L35-L36" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "standalone_worker_v2.py", "first": 197, "last": 227, "engine": null, "section": "inventory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/standalone_worker_v2.py", "reviewed_symbol": "_validate_vocab_compatibility", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/standalone_worker_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/standalone_worker_v2.py", "symbol": "_validate_vocab_compatibility", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/standalone_worker_v2.py#L197-L227" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 2312, "last": 2317, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2312-L2317" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/decoupled_spec_io.py", "first": 43, "last": 50, "engine": "sglang", "section": "inventory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/decoupled_spec_io.py#L43-L50" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 290, "last": 296, "engine": "vllm", "section": "verification-rules", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L290-L296" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 2192, "last": 2196, "engine": "sglang", "section": "verification-rules", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2192-L2196" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "server_args.py", "first": 2182, "last": 2191, "engine": null, "section": "verification-rules", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "speculative_accept_threshold_single", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "speculative_accept_threshold_single", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2182-L2191" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 219, "last": 225, "engine": "vllm", "section": "verification-rules", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L219-L225" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/rejection_sampler_utils.py", "first": 595, "last": 626, "engine": "vllm", "section": "verification-rules", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/rejection_sampler_utils.py#L595-L626" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/rejection_sampler.py", "first": 84, "last": 97, "engine": "vllm", "section": "verification-rules", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/rejection_sampler.py#L84-L97" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 77, "last": 90, "engine": "vllm", "section": "verification-rules", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L77-L90" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 245, "last": 253, "engine": "vllm", "section": "verification-rules", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L245-L253" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/spec_info.py", "first": 72, "last": 87, "engine": "sglang", "section": "registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_info.py#L72-L87" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/spec_registry.py", "first": 206, "last": 219, "engine": "sglang", "section": "registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_registry.py#L206-L219" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "spec_registry.py", "first": 174, "last": 184, "engine": null, "section": "registry", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/spec_registry.py", "reviewed_symbol": "_reserved_names", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/spec_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/spec_registry.py", "symbol": "_reserved_names", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_registry.py#L174-L184" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 634, "last": 651, "engine": "vllm", "section": "registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L634-L651" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 1496, "last": 1496, "engine": "vllm", "section": "registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L1496-L1496" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/spec_decode/custom_class_proposer.py", "first": 12, "last": 21, "engine": "vllm", "section": "registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/custom_class_proposer.py#L12-L21" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 729, "last": 739, "engine": "vllm", "section": "registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L729-L739" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dynamic/utils.py", "first": 77, "last": 88, "engine": null, "section": "adaptive", "declared_engine": "vllm", "declared_path": "vllm/v1/spec_decode/dynamic/utils.py", "reviewed_symbol": "build_dynamic_sd_schedule_lookup", "resolved_engine": "vllm", "resolved_path": "vllm/v1/spec_decode/dynamic/utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/spec_decode/dynamic/utils.py", "symbol": "build_dynamic_sd_schedule_lookup", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/dynamic/utils.py#L77-L88" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/eagle_worker_v2.py", "first": 1152, "last": 1152, "engine": "sglang", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_v2.py#L1152-L1152" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "adaptive_spec_params.py", "first": 329, "last": 345, "engine": null, "section": "adaptive", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/adaptive_spec_params.py", "reviewed_symbol": "_route", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/adaptive_spec_params.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "symbol": "_route", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L329-L345" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/managers/scheduler_components/batch_result_processor.py", "first": 652, "last": 657, "engine": "sglang", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/batch_result_processor.py#L652-L657" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "adaptive_spec_params.py", "first": 174, "last": 198, "engine": null, "section": "adaptive", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/adaptive_spec_params.py", "reviewed_symbol": "update", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/adaptive_spec_params.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "symbol": "update", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L174-L198" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "first": 217, "last": 236, "engine": "sglang", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L217-L236" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/adaptive_runtime_state.py", "first": 18, "last": 27, "engine": "sglang", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_runtime_state.py#L18-L27" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "eagle_worker_v2.py", "first": 1386, "last": 1425, "engine": null, "section": "adaptive", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_worker_v2.py", "reviewed_symbol": "apply_runtime_state", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_worker_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_worker_v2.py", "symbol": "apply_runtime_state", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_worker_v2.py#L1386-L1425" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "adaptive_runtime_state.py", "first": 94, "last": 111, "engine": null, "section": "adaptive", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/adaptive_runtime_state.py", "reviewed_symbol": "init_states", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/adaptive_runtime_state.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/adaptive_runtime_state.py", "symbol": "init_states", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_runtime_state.py#L94-L111" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 241, "last": 243, "engine": "vllm", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L241-L243" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/adaptive_verification.py", "first": 41, "last": 60, "engine": "vllm", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/adaptive_verification.py#L41-L60" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/adaptive_verification.py", "first": 310, "last": 330, "engine": "vllm", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/adaptive_verification.py#L310-L330" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/dspark_components/dspark_planner.py", "first": 965, "last": 972, "engine": "sglang", "section": "adaptive", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dspark_components/dspark_planner.py#L965-L972" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "adaptive_verification.py", "first": 86, "last": 90, "engine": null, "section": "adaptive", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu/spec_decode/adaptive_verification.py", "reviewed_symbol": "values", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/spec_decode/adaptive_verification.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu/spec_decode/adaptive_verification.py", "symbol": "values", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/adaptive_verification.py#L86-L90" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/spec_info.py", "first": 151, "last": 154, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_info.py#L151-L154" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/eagle_disaggregation.py", "first": 90, "last": 97, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_disaggregation.py#L90-L97" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "eagle_disaggregation.py", "first": 27, "last": 31, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/eagle_disaggregation.py", "reviewed_symbol": "get_spec", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/eagle_disaggregation.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/eagle_disaggregation.py", "symbol": "get_spec", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/eagle_disaggregation.py#L27-L31" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dflash_disaggregation.py", "first": 21, "last": 24, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/dflash_disaggregation.py", "reviewed_symbol": "spec_info", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/dflash_disaggregation.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/dflash_disaggregation.py", "symbol": "spec_info", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_disaggregation.py#L21-L24" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dspark_disaggregation.py", "first": 21, "last": 24, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/dspark_disaggregation.py", "reviewed_symbol": "make_next_draft_input", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/dspark_disaggregation.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/dspark_disaggregation.py", "symbol": "make_next_draft_input", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dspark_disaggregation.py#L21-L24" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "dflash_worker_v2.py", "first": 484, "last": 484, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/dflash_worker_v2.py", "reviewed_symbol": "lm_head", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/dflash_worker_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/dflash_worker_v2.py", "symbol": "lm_head", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/dflash_worker_v2.py#L484-L484" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/lora/utils.py", "first": 492, "last": 502, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/lora/utils.py#L492-L502" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 322, "last": 324, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L322-L324" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/environ.py", "first": 1137, "last": 1137, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1137-L1137" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 9590, "last": 9601, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9590-L9601" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/lora/utils.py", "first": 25, "last": 32, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/lora/utils.py#L25-L32" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/config/speculative.py", "first": 977, "last": 977, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L977-L977" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/spec_decode/metrics.py", "first": 48, "last": 49, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/metrics.py#L48-L49" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 2152, "last": 2160, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2152-L2160" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "adaptive_spec_params.py", "first": 253, "last": 257, "engine": null, "section": "hands-on", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/adaptive_spec_params.py", "reviewed_symbol": "log_info_on_rank0", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/adaptive_spec_params.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "symbol": "log_info_on_rank0", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L253-L257" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/speculative/adaptive_spec_params.py", "first": 200, "last": 248, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/adaptive_spec_params.py#L200-L248" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "python/sglang/srt/server_args.py", "first": 9563, "last": 9601, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9563-L9601" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/adaptive_verification.py", "first": 41, "last": 60, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/adaptive_verification.py#L41-L60" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "speculative_hook.py", "first": 45, "last": 58, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/arg_groups/speculative_hook.py", "reviewed_symbol": "speculative_algorithm", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/arg_groups/speculative_hook.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "symbol": "speculative_algorithm", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L45-L58" }, { "page": "06-decoding-algorithms/06-06-speculative-decoding-zoo.html", "path": "vllm/v1/worker/gpu/spec_decode/rejection_sampler_utils.py", "first": 596, "last": 596, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/rejection_sampler_utils.py#L596-L596" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "benchmarks/kernels/benchmark_moe_defaults.py", "first": 156, "last": 175, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/kernels/benchmark_moe_defaults.py#L156-L175" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/configs/model_config.py", "first": 411, "last": 420, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L411-L420" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "tests/kernels/moe/test_routing.py", "first": 83, "last": 84, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/kernels/moe/test_routing.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/kernels/moe/test_routing.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/kernels/moe/test_routing.py#L83-L84" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "test/registered/moe/test_topk_renormalize_degenerate.py", "first": 102, "last": 103, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "test/registered/moe/test_topk_renormalize_degenerate.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "test/registered/moe/test_topk_renormalize_degenerate.py", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/registered/moe/test_topk_renormalize_degenerate.py#L102-L103" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/unquantized_fused_moe_method.py", "first": 64, "last": 95, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/unquantized_fused_moe_method.py#L64-L95" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/router/fused_topk_router.py", "first": 80, "last": 100, "engine": "vllm", "section": "routing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/router/fused_topk_router.py#L80-L100" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/models/deepseek_v2.py", "first": 370, "last": 386, "engine": "vllm", "section": "routing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_v2.py#L370-L386" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py", "first": 112, "last": 161, "engine": "vllm", "section": "routing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py#L112-L161" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/topk.py", "first": 1352, "last": 1361, "engine": "sglang", "section": "routing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/topk.py#L1352-L1361" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/topk.py", "first": 1309, "last": 1350, "engine": "sglang", "section": "routing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/topk.py#L1309-L1350" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/topk.py", "first": 215, "last": 223, "engine": "sglang", "section": "routing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/topk.py#L215-L223" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/oracle/fp8.py", "first": 112, "last": 123, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/oracle/fp8.py#L112-L123" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/config/kernel.py", "first": 193, "last": 193, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/kernel.py#L193-L193" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/engine/arg_utils.py", "first": 1621, "last": 1623, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L1621-L1623" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/envs.py", "first": 197, "last": 198, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L197-L198" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/server_args.py", "first": 2374, "last": 2400, "engine": "sglang", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2374-L2400" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/quantization/fp8.py", "first": 1096, "last": 1119, "engine": "sglang", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/fp8.py#L1096-L1119" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/deep_gemm_wrapper/configurer.py", "first": 17, "last": 35, "engine": "sglang", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/deep_gemm_wrapper/configurer.py#L17-L35" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/moe_align_block_size.py", "first": 59, "last": 72, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/moe_align_block_size.py#L59-L72" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/moe_runner/triton_utils/moe_align_block_size.py", "first": 60, "last": 73, "engine": "sglang", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/moe_runner/triton_utils/moe_align_block_size.py#L60-L73" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "csrc/libtorch_stable/moe/moe_align_sum_kernels.cu", "first": 160, "last": 194, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/moe/moe_align_sum_kernels.cu#L160-L194" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 404, "last": 423, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L404-L423" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1854, "last": 1857, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1854-L1857" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1557, "last": 1570, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1557-L1570" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/moe_runner/deep_gemm.py", "first": 881, "last": 896, "engine": "sglang", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/moe_runner/deep_gemm.py#L881-L896" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/moe_runner/deep_gemm.py", "first": 951, "last": 1028, "engine": "sglang", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/moe_runner/deep_gemm.py#L951-L1028" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1320, "last": 1324, "engine": "vllm", "section": "grouped-gemm", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1320-L1324" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/transformers_utils/configs/arctic.py", "first": 150, "last": 151, "engine": "vllm", "section": "capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/transformers_utils/configs/arctic.py#L150-L151" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/moe_align_block_size.py", "first": 74, "last": 74, "engine": "vllm", "section": "capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/moe_align_block_size.py#L74-L74" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/layer.py", "first": 259, "last": 259, "engine": "vllm", "section": "capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/layer.py#L259-L259" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/environ.py", "first": 989, "last": 989, "engine": "sglang", "section": "capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L989-L989" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/kernels/ops/moe/ep_moe_kernels.py", "first": 1547, "last": 1576, "engine": "sglang", "section": "capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/moe/ep_moe_kernels.py#L1547-L1576" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/moe_runner/deep_gemm.py", "first": 117, "last": 152, "engine": "sglang", "section": "capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/moe_runner/deep_gemm.py#L117-L152" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/models/qwen3_moe.py", "first": 175, "last": 197, "engine": "vllm", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/qwen3_moe.py#L175-L197" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/models/deepseek_v2.py", "first": 584, "last": 604, "engine": "sglang", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_v2.py#L584-L604" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/models/deepseek_v2.py", "first": 391, "last": 394, "engine": "vllm", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_v2.py#L391-L394" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/utils.py", "first": 73, "last": 85, "engine": "vllm", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/utils.py#L73-L85" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/_aiter_ops.py", "first": 195, "last": 207, "engine": "vllm", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/_aiter_ops.py#L195-L207" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/models/deepseek_v2.py", "first": 355, "last": 368, "engine": "vllm", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_v2.py#L355-L368" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/server_args.py", "first": 2543, "last": 2550, "engine": "sglang", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2543-L2550" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/models/deepseek_v2.py", "first": 3019, "last": 3074, "engine": "sglang", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_v2.py#L3019-L3074" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/utils.py", "first": 496, "last": 505, "engine": "sglang", "section": "shared-experts", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/utils.py#L496-L505" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/quantization/moe_wna16.py", "first": 171, "last": 197, "engine": "vllm", "section": "memory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/moe_wna16.py#L171-L197" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/models/deepseek_v2.py", "first": 411, "last": 421, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_v2.py#L411-L421" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py", "first": 112, "last": 161, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py#L112-L161" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/moe_runner/deep_gemm.py", "first": 608, "last": 617, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/moe_runner/deep_gemm.py#L608-L617" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/model_executor/layers/fused_moe/layer.py", "first": 247, "last": 252, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/layer.py#L247-L252" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/config/parallel.py", "first": 72, "last": 72, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L72-L72" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "grouped_topk_router.py", "first": 133, "last": 134, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1629, "page": "07-architectures/07-01-mixture-of-experts.html", "section": "routing", "engine": "vllm", "path": "vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py", "first": 112, "last": 161 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py#L133-L134" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "topk.py", "first": 1331, "last": 1336, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/moe/topk.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1631, "page": "07-architectures/07-01-mixture-of-experts.html", "section": "routing", "engine": "sglang", "path": "python/sglang/srt/layers/moe/topk.py", "first": 1309, "last": 1350 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/topk.py#L1331-L1336" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/envs.py", "first": 91, "last": 91, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L91-L91" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "vllm/config/parallel.py", "first": 165, "last": 165, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L165-L165" }, { "page": "07-architectures/07-01-mixture-of-experts.html", "path": "python/sglang/srt/layers/moe/moe_runner/deep_gemm.py", "first": 955, "last": 966, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/moe_runner/deep_gemm.py#L955-L966" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 2082, "last": 2097, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L2082-L2097" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 13, "last": 21, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L13-L21" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "tests/v1/metrics/test_perf_metrics.py", "first": 1030, "last": 1041, "engine": "vllm", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/metrics/test_perf_metrics.py#L1030-L1041" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mha.py", "first": 83, "last": 88, "engine": "sglang", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mha.py#L83-L88" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 36, "last": 64, "engine": "vllm", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L36-L64" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "tests/v1/metrics/test_perf_metrics.py", "first": 1035, "last": 1035, "engine": null, "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/metrics/test_perf_metrics.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/metrics/test_perf_metrics.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/metrics/test_perf_metrics.py#L1035-L1035" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "tests/v1/metrics/test_perf_metrics.py", "first": 1034, "last": 1034, "engine": null, "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/metrics/test_perf_metrics.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/metrics/test_perf_metrics.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/metrics/test_perf_metrics.py#L1034-L1034" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "tests/v1/metrics/test_perf_metrics.py", "first": 1033, "last": 1033, "engine": null, "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/metrics/test_perf_metrics.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/metrics/test_perf_metrics.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/metrics/test_perf_metrics.py#L1033-L1033" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 39, "last": 39, "engine": "vllm", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L39-L39" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 41, "last": 41, "engine": "vllm", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L41-L41" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 40, "last": 40, "engine": "vllm", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L40-L40" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 42, "last": 42, "engine": "vllm", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L42-L42" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 38, "last": 38, "engine": "vllm", "section": "the-shapes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L38-L38" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/mla.py", "first": 189, "last": 192, "engine": "vllm", "section": "why-rope-is-separate", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mla.py#L189-L192" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 940, "last": 956, "engine": "vllm", "section": "absorbed-weight-trick", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L940-L956" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 1101, "last": 1170, "engine": "vllm", "section": "absorbed-weight-trick", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L1101-L1170" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_common/deepseek_weight_loader.py", "first": 653, "last": 655, "engine": "sglang", "section": "absorbed-weight-trick", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_common/deepseek_weight_loader.py#L653-L655" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/v1/attention/backends/mla/triton_mla.py", "first": 286, "last": 320, "engine": "vllm", "section": "absorbed-weight-trick", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/triton_mla.py#L286-L320" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mla.py", "first": 722, "last": 728, "engine": "sglang", "section": "absorbed-weight-trick", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mla.py#L722-L728" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 1215, "last": 1247, "engine": "vllm", "section": "absorbed-weight-trick", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L1215-L1247" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_v2.py", "first": 1896, "last": 1928, "engine": "sglang", "section": "prefill-vs-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_v2.py#L1896-L1928" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_common/attention_backend_handler.py", "first": 99, "last": 132, "engine": "sglang", "section": "prefill-vs-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_common/attention_backend_handler.py#L99-L132" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/environ.py", "first": 590, "last": 590, "engine": "sglang", "section": "prefill-vs-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L590-L590" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mha.py", "first": 96, "last": 101, "engine": "sglang", "section": "prefill-vs-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mha.py#L96-L101" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 821, "last": 880, "engine": "vllm", "section": "prefill-vs-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L821-L880" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/v1/attention/backends/mla/flashattn_mla.py", "first": 117, "last": 118, "engine": "vllm", "section": "prefill-vs-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/flashattn_mla.py#L117-L118" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/v1/attention/backends/mla/flashmla.py", "first": 120, "last": 122, "engine": "vllm", "section": "prefill-vs-decode", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/flashmla.py#L120-L122" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/v1/attention/backends/registry.py", "first": 44, "last": 130, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L44-L130" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "triton_mla.py", "first": 149, "last": 150, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/mla/triton_mla.py", "reviewed_symbol": "supports_compute_capability", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/mla/triton_mla.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/mla/triton_mla.py", "symbol": "supports_compute_capability", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/triton_mla.py#L149-L150" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "flashattn_mla.py", "first": 67, "last": 69, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/mla/flashattn_mla.py", "reviewed_symbol": "supports_batch_invariance", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/mla/flashattn_mla.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/mla/flashattn_mla.py", "symbol": "supports_batch_invariance", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/flashattn_mla.py#L67-L69" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "flashmla_sparse.py", "first": 91, "last": 95, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/mla/flashmla_sparse.py", "reviewed_symbol": "supported_kv_cache_dtypes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/mla/flashmla_sparse.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/mla/flashmla_sparse.py", "symbol": "supported_kv_cache_dtypes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/flashmla_sparse.py#L91-L95" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "flashinfer_mla_sparse.py", "first": 145, "last": 150, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/mla/flashinfer_mla_sparse.py", "reviewed_symbol": "supported_kv_cache_dtypes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/mla/flashinfer_mla_sparse.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/mla/flashinfer_mla_sparse.py", "symbol": "supported_kv_cache_dtypes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/flashinfer_mla_sparse.py#L145-L150" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 1061, "last": 1067, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L1061-L1067" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/platforms/cuda.py", "first": 93, "last": 143, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L93-L143" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/layers/attention/attention_registry.py", "first": 42, "last": 66, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L42-L66" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/mla.py", "first": 170, "last": 192, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mla.py#L170-L192" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/mla.py", "first": 226, "last": 226, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mla.py#L226-L226" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 359, "last": 413, "engine": "sglang", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L359-L413" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 4412, "last": 4412, "engine": "sglang", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L4412-L4412" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 4366, "last": 4367, "engine": "sglang", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L4366-L4367" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/v1/attention/backends/mla/indexer.py", "first": 481, "last": 481, "engine": "vllm", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mla/indexer.py#L481-L481" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/models/deepseek_v2.py", "first": 648, "last": 719, "engine": "vllm", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_v2.py#L648-L719" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/model_executor/pool_configurator.py", "first": 377, "last": 409, "engine": "sglang", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/pool_configurator.py#L377-L409" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/models/deepseek_v2.py", "first": 1119, "last": 1145, "engine": "vllm", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/deepseek_v2.py#L1119-L1145" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mla.py", "first": 722, "last": 735, "engine": "sglang", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_common/attention_forward_methods/forward_mla.py#L722-L735" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/mem_cache/deepseek_v4_memory_pool.py", "first": 34, "last": 48, "engine": "sglang", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/deepseek_v4_memory_pool.py#L34-L48" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/v1/kv_cache_interface.py", "first": 385, "last": 400, "engine": "vllm", "section": "dsa-indexer", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L385-L400" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/config/parallel.py", "first": 342, "last": 342, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L342-L342" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/server_args.py", "first": 1033, "last": 1040, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1033-L1040" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 1091, "last": 1100, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L1091-L1100" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/v1/metrics/perf.py", "first": 677, "last": 678, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/perf.py#L677-L678" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "vllm/model_executor/layers/attention/mla_attention.py", "first": 94, "last": 118, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/mla_attention.py#L94-L118" }, { "page": "07-architectures/07-02-mla-in-depth.html", "path": "python/sglang/srt/models/deepseek_common/attention_backend_handler.py", "first": 186, "last": 195, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/deepseek_common/attention_backend_handler.py#L186-L195" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/platforms/interface.py", "first": 899, "last": 914, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/interface.py#L899-L914" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/config/cache.py", "first": 68, "last": 79, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L68-L79" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 699, "last": 709, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L699-L709" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/layers/mamba/mamba_utils.py", "first": 194, "last": 208, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mamba/mamba_utils.py#L194-L208" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/transformers_utils/configs/nemotron_h.py", "first": 176, "last": 183, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/transformers_utils/configs/nemotron_h.py#L176-L183" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/models/nemotron_h.py", "first": 371, "last": 371, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/nemotron_h.py#L371-L371" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/models/config.py", "first": 651, "last": 652, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/config.py#L651-L652" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/attention/selector.py", "first": 154, "last": 190, "engine": "vllm", "section": "second-registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L154-L190" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/attention/backends/registry.py", "first": 176, "last": 192, "engine": "vllm", "section": "second-registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L176-L192" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/attention/selector.py", "first": 229, "last": 242, "engine": "vllm", "section": "second-registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L229-L242" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/layers/mamba/mamba_mixer2.py", "first": 1146, "last": 1148, "engine": "vllm", "section": "second-registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mamba/mamba_mixer2.py#L1146-L1148" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/layers/mamba/short_conv.py", "first": 331, "last": 333, "engine": "vllm", "section": "second-registry", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mamba/short_conv.py#L331-L333" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 546, "last": 577, "engine": "sglang", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L546-L577" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/attention/backends/utils.py", "first": 1019, "last": 1033, "engine": "vllm", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/utils.py#L1019-L1033" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/models/config.py", "first": 646, "last": 647, "engine": "vllm", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/config.py#L646-L647" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/kv_cache_interface.py", "first": 687, "last": 696, "engine": "vllm", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L687-L696" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/layers/mamba/abstract.py", "first": 63, "last": 71, "engine": "vllm", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mamba/abstract.py#L63-L71" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/config/cache.py", "first": 40, "last": 46, "engine": "vllm", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L40-L46" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 1400, "last": 1429, "engine": "vllm", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L1400-L1429" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/platforms/interface.py", "first": 889, "last": 940, "engine": "vllm", "section": "allocation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/interface.py#L889-L940" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 159, "last": 167, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L159-L167" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/mamba_radix_cache.py", "first": 1209, "last": 1214, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/mamba_radix_cache.py#L1209-L1214" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/mamba_radix_cache.py", "first": 84, "last": 96, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/mamba_radix_cache.py#L84-L96" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/mamba_radix_cache.py", "first": 1100, "last": 1127, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/mamba_radix_cache.py#L1100-L1127" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/mamba_radix_cache.py", "first": 1181, "last": 1193, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/mamba_radix_cache.py#L1181-L1193" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/server_args.py", "first": 2620, "last": 2624, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2620-L2624" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/server_args.py", "first": 6280, "last": 6300, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L6280-L6300" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/mamba_checkpoint_pool.py", "first": 18, "last": 28, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/mamba_checkpoint_pool.py#L18-L28" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/config/cache.py", "first": 157, "last": 164, "engine": "vllm", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L157-L164" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 1347, "last": 1371, "engine": "vllm", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L1347-L1371" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/config/cache.py", "first": 107, "last": 107, "engine": "vllm", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L107-L107" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/models/config.py", "first": 612, "last": 620, "engine": "vllm", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/config.py#L612-L620" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 1567, "last": 1580, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L1567-L1580" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 1495, "last": 1515, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L1495-L1515" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/server_args.py", "first": 5792, "last": 5798, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5792-L5798" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 1520, "last": 1546, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L1520-L1546" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 1580, "last": 1591, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L1580-L1591" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/server_args.py", "first": 2571, "last": 2578, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2571-L2578" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/runtime_context.py", "first": 1479, "last": 1490, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/runtime_context.py#L1479-L1490" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/server_args.py", "first": 9177, "last": 9199, "engine": "sglang", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9177-L9199" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 324, "last": 334, "engine": "vllm", "section": "prefix-reuse", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L324-L334" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1347, "last": 1388, "engine": "vllm", "section": "preemption-chunking", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1347-L1388" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1698, "last": 1706, "engine": "sglang", "section": "preemption-chunking", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1698-L1706" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1720, "last": 1731, "engine": "sglang", "section": "preemption-chunking", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1720-L1731" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/attention/backends/mamba_attn.py", "first": 568, "last": 570, "engine": "vllm", "section": "preemption-chunking", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mamba_attn.py#L568-L570" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/attention/backends/mamba2_attn.py", "first": 145, "last": 156, "engine": "vllm", "section": "preemption-chunking", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mamba2_attn.py#L145-L156" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 381, "last": 388, "engine": "vllm", "section": "preemption-chunking", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L381-L388" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/layers/attention/mamba/mamba.py", "first": 600, "last": 606, "engine": "sglang", "section": "kernels", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/mamba/mamba.py#L600-L606" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/layers/attention/mamba/mamba.py", "first": 739, "last": 752, "engine": "sglang", "section": "kernels", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/mamba/mamba.py#L739-L752" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/layers/attention/hybrid_linear_attn_backend.py", "first": 1090, "last": 1107, "engine": "sglang", "section": "kernels", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/hybrid_linear_attn_backend.py#L1090-L1107" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "single_type_kv_cache_manager.py", "first": 1295, "last": 1371, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/core/single_type_kv_cache_manager.py", "reviewed_symbol": "find_longest_cache_hit", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/single_type_kv_cache_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "symbol": "find_longest_cache_hit", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L1295-L1371" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/core/kv_cache_coordinator.py", "first": 800, "last": 845, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_coordinator.py#L800-L845" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "mamba2_attn.py", "first": 145, "last": 156, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/mamba2_attn.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1773, "page": "07-architectures/07-03-ssm-and-hybrids.html", "section": "preemption-chunking", "engine": "vllm", "path": "vllm/v1/attention/backends/mamba2_attn.py", "first": 145, "last": 156 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/mamba2_attn.py#L145-L156" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/model_executor/models/config.py", "first": 631, "last": 634, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/config.py#L631-L634" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "single_type_kv_cache_manager.py", "first": 1461, "last": 1465, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/core/single_type_kv_cache_manager.py", "reviewed_symbol": "get_num_common_prefix_blocks", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/single_type_kv_cache_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "symbol": "get_num_common_prefix_blocks", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L1461-L1465" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 806, "last": 831, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L806-L831" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "vllm/v1/kv_cache_interface.py", "first": 698, "last": 709, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L698-L709" }, { "page": "07-architectures/07-03-ssm-and-hybrids.html", "path": "python/sglang/srt/mem_cache/mamba_radix_cache.py", "first": 1170, "last": 1179, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/mamba_radix_cache.py#L1170-L1179" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1648, "last": 1667, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1648-L1667" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/config/scheduler.py", "first": 86, "last": 97, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L86-L97" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/config/scheduler.py", "first": 238, "last": 239, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L238-L239" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/model_executor/models/utils.py", "first": 701, "last": 713, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/utils.py#L701-L713" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/model_executor/models/qwen2_vl.py", "first": 916, "last": 923, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/qwen2_vl.py#L916-L923" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/multimodal/processors/qwen_vl.py", "first": 130, "last": 140, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/multimodal/processors/qwen_vl.py#L130-L140" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/multimodal/processors/qwen_vl.py", "first": 47, "last": 50, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/multimodal/processors/qwen_vl.py#L47-L50" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/environ.py", "first": 1159, "last": 1159, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1159-L1159" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/encoder_cache_manager.py", "first": 117, "last": 127, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/encoder_cache_manager.py#L117-L127" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/encoder_cache_manager.py", "first": 166, "last": 187, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/encoder_cache_manager.py#L166-L187" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/encoder_cache_manager.py", "first": 322, "last": 327, "engine": "vllm", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/encoder_cache_manager.py#L322-L327" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/mem_cache/multimodal_cache.py", "first": 102, "last": 121, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/multimodal_cache.py#L102-L121" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/environ.py", "first": 1158, "last": 1158, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1158-L1158" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/mem_cache/kv_cache_builder.py", "first": 374, "last": 375, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_builder.py#L374-L375" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/mm_schedule.py", "first": 512, "last": 529, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/mm_schedule.py#L512-L529" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/mm_schedule.py", "first": 229, "last": 246, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/mm_schedule.py#L229-L246" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/mm_schedule.py", "first": 285, "last": 355, "engine": "sglang", "section": "how-production-systems-do-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/mm_schedule.py#L285-L355" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/config/cache.py", "first": 107, "last": 107, "engine": "vllm", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L107-L107" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/server_args.py", "first": 952, "last": 954, "engine": "sglang", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L952-L954" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 513, "last": 523, "engine": "vllm", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L513-L523" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 608, "last": 610, "engine": "vllm", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L608-L610" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/processing/inputs.py", "first": 67, "last": 75, "engine": "vllm", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/processing/inputs.py#L67-L75" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/hasher.py", "first": 22, "last": 47, "engine": "vllm", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/hasher.py#L22-L47" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/hasher.py", "first": 62, "last": 76, "engine": "vllm", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/hasher.py#L62-L76" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 146, "last": 149, "engine": "sglang", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L146-L149" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/mm_utils.py", "first": 314, "last": 318, "engine": "sglang", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/mm_utils.py#L314-L318" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/models/mimo_v2_nextn.py", "first": 202, "last": 202, "engine": "sglang", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/mimo_v2_nextn.py#L202-L202" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/models/kimi_k25_eagle3.py", "first": 265, "last": 265, "engine": "sglang", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/kimi_k25_eagle3.py#L265-L265" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 196, "last": 202, "engine": "sglang", "section": "prefix-caching-with-images", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L196-L202" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1632, "last": 1647, "engine": "vllm", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1632-L1647" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/encoder_cache_manager.py", "first": 310, "last": 320, "engine": "vllm", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/encoder_cache_manager.py#L310-L320" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/mm_schedule.py", "first": 28, "last": 70, "engine": "sglang", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/mm_schedule.py#L28-L70" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/encoder_cudagraph.py", "first": 190, "last": 201, "engine": "vllm", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/encoder_cudagraph.py#L190-L201" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/model_executor/models/qwen2_5_vl.py", "first": 1758, "last": 1762, "engine": "vllm", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/qwen2_5_vl.py#L1758-L1762" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/config/compilation.py", "first": 543, "last": 543, "engine": "vllm", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L543-L543" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 6741, "last": 6741, "engine": "vllm", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L6741-L6741" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/config/compilation.py", "first": 548, "last": 556, "engine": "vllm", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L548-L556" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/environ.py", "first": 1163, "last": 1163, "engine": "sglang", "section": "scheduling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1163-L1163" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/disaggregation/encoder/receiver.py", "first": 397, "last": 432, "engine": "sglang", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/encoder/receiver.py#L397-L432" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/disaggregation/encoder/receiver.py", "first": 1493, "last": 1503, "engine": "sglang", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/encoder/receiver.py#L1493-L1503" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/server_args.py", "first": 336, "last": 343, "engine": "sglang", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L336-L343" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/disaggregation/encoder/receiver.py", "first": 2430, "last": 2470, "engine": "sglang", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/encoder/receiver.py#L2430-L2470" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/server_args.py", "first": 3232, "last": 3241, "engine": "sglang", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3232-L3241" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/server_args.py", "first": 7809, "last": 7812, "engine": "sglang", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L7809-L7812" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/server_args.py", "first": 7850, "last": 7869, "engine": "sglang", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L7850-L7869" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/distributed/ec_transfer/ec_connector/base.py", "first": 1, "last": 24, "engine": "vllm", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/ec_transfer/ec_connector/base.py#L1-L24" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1683, "last": 1689, "engine": "vllm", "section": "epd", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1683-L1689" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/processing/processor.py", "first": 1839, "last": 1870, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/processing/processor.py#L1839-L1870" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/processing/inputs.py", "first": 67, "last": 75, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/processing/inputs.py#L67-L75" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/registry.py", "first": 275, "last": 299, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/registry.py#L275-L299" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/config/multimodal.py", "first": 164, "last": 164, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/multimodal.py#L164-L164" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/cache.py", "first": 444, "last": 458, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/cache.py#L444-L458" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/cache.py", "first": 392, "last": 410, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/cache.py#L392-L410" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/config/multimodal.py", "first": 152, "last": 152, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/multimodal.py#L152-L152" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 592, "last": 607, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L592-L607" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 730, "last": 738, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L730-L738" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3077, "last": 3080, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3077-L3080" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3283, "last": 3291, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3283-L3291" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3352, "last": 3363, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3352-L3363" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/model_executor/models/interfaces.py", "first": 453, "last": 487, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/interfaces.py#L453-L487" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 745, "last": 757, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L745-L757" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 1238, "last": 1244, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L1238-L1244" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/mm_schedule.py", "first": 615, "last": 617, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/mm_schedule.py#L615-L617" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/managers/mm_utils.py", "first": 286, "last": 290, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/mm_utils.py#L286-L290" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/multimodal/processors/base_processor.py", "first": 295, "last": 355, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/multimodal/processors/base_processor.py#L295-L355" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/multimodal/processors/executor.py", "first": 15, "last": 26, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/multimodal/processors/executor.py#L15-L26" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/multimodal/processors/qwen_vl.py", "first": 317, "last": 323, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/multimodal/processors/qwen_vl.py#L317-L323" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "python/sglang/srt/multimodal/processors/llava.py", "first": 138, "last": 145, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/multimodal/processors/llava.py#L138-L145" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3124, "last": 3130, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3124-L3130" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3211, "last": 3224, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3211-L3224" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1622, "last": 1625, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1622-L1625" }, { "page": "07-architectures/07-04-multimodal-serving.html", "path": "vllm/multimodal/hasher.py", "first": 35, "last": 35, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/multimodal/hasher.py#L35-L35" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/worker_manager.py", "first": 272, "last": 285, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/worker_manager.py#L272-L285" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/layers/base_linear.py", "first": 129, "last": 150, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/layers/base_linear.py#L129-L150" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/layers/base_linear.py", "first": 204, "last": 208, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/layers/base_linear.py#L204-L208" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/platforms/cuda.py", "first": 560, "last": 562, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L560-L562" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/punica_wrapper/punica_gpu.py", "first": 245, "last": 267, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/punica_wrapper/punica_gpu.py#L245-L267" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/ops/triton_ops/lora_shrink_op.py", "first": 133, "last": 143, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/ops/triton_ops/lora_shrink_op.py#L133-L143" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/ops/triton_ops/lora_kernel_metadata.py", "first": 135, "last": 147, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/ops/triton_ops/lora_kernel_metadata.py#L135-L147" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/ops/triton_ops/lora_shrink_op.py", "first": 70, "last": 95, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/ops/triton_ops/lora_shrink_op.py#L70-L95" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/config/lora.py", "first": 35, "last": 46, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/lora.py#L35-L46" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/engine/arg_utils.py", "first": 611, "last": 616, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L611-L616" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/config/compilation.py", "first": 660, "last": 667, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L660-L667" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/config/vllm.py", "first": 2496, "last": 2500, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L2496-L2500" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/lora/backend/chunked_backend.py", "first": 25, "last": 36, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/lora/backend/chunked_backend.py#L25-L36" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/kernels/ops/gemm/chunked_sgmv_shrink.py", "first": 69, "last": 79, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/gemm/chunked_sgmv_shrink.py#L69-L79" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "lora_shrink_op.py", "first": 197, "last": 197, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/lora/ops/triton_ops/lora_shrink_op.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1919, "page": "07-architectures/07-05-lora-and-pooling-models.html", "section": "exercises", "engine": "vllm", "path": "vllm/lora/ops/triton_ops/lora_shrink_op.py", "first": 196, "last": 198 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/ops/triton_ops/lora_shrink_op.py#L197-L197" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/layers/base_linear.py", "first": 111, "last": 120, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/layers/base_linear.py#L111-L120" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/server_args.py", "first": 2925, "last": 2929, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2925-L2929" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/server_args.py", "first": 2953, "last": 2965, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2953-L2965" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/utils.py", "first": 79, "last": 96, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/utils.py#L79-L96" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/utils/common.py", "first": 4328, "last": 4351, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L4328-L4351" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/model_manager.py", "first": 315, "last": 337, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/model_manager.py#L315-L337" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/model_manager.py", "first": 1221, "last": 1233, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/model_manager.py#L1221-L1233" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/lora/mem_pool.py", "first": 752, "last": 779, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/lora/mem_pool.py#L752-L779" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/lora/lora_manager.py", "first": 368, "last": 373, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/lora/lora_manager.py#L368-L373" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/server_args.py", "first": 9446, "last": 9456, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9446-L9456" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3535, "last": 3546, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3535-L3546" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/server_args.py", "first": 9563, "last": 9583, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9563-L9583" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/speculative/spec_info.py", "first": 131, "last": 136, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_info.py#L131-L136" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py", "first": 350, "last": 358, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py#L350-L358" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/environ.py", "first": 1137, "last": 1137, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1137-L1137" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/server_args.py", "first": 9590, "last": 9593, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9590-L9593" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/speculative/ragged_verify.py", "first": 13, "last": 16, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/ragged_verify.py#L13-L16" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "server_args.py", "first": 9596, "last": 9601, "engine": null, "section": "production", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "speculative_adaptive", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "speculative_adaptive", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9596-L9601" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/v1/worker/lora_model_runner_mixin.py", "first": 64, "last": 84, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/lora_model_runner_mixin.py#L64-L84" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 539, "last": 551, "engine": "vllm", "section": "cache-partition", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L539-L551" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 935, "last": 940, "engine": "sglang", "section": "cache-partition", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L935-L940" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/request.py", "first": 30, "last": 30, "engine": "vllm", "section": "cache-partition", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/request.py#L30-L30" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/lora/lora_registry.py", "first": 37, "last": 37, "engine": "sglang", "section": "cache-partition", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/lora/lora_registry.py#L37-L37" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/model_executor/layers/pooler/seqwise/methods.py", "first": 37, "last": 58, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/pooler/seqwise/methods.py#L37-L58" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/layers/pooler.py", "first": 57, "last": 74, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/pooler.py#L57-L74" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/config/model.py", "first": 1169, "last": 1191, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1169-L1191" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/server_args.py", "first": 600, "last": 602, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L600-L602" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/config/model.py", "first": 2040, "last": 2083, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L2040-L2083" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/engine/arg_utils.py", "first": 2673, "last": 2723, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2673-L2723" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1218, "last": 1224, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1218-L1224" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/v1/kv_cache_interface.py", "first": 720, "last": 724, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L720-L724" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 1444, "last": 1449, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L1444-L1449" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "first": 1265, "last": 1269, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L1265-L1269" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/models/bert.py", "first": 247, "last": 255, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/bert.py#L247-L255" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3534, "last": 3553, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3534-L3553" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/tasks.py", "first": 10, "last": 19, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tasks.py#L10-L19" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/tasks.py", "first": 33, "last": 38, "engine": "vllm", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tasks.py#L33-L38" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/layers/pooler.py", "first": 233, "last": 263, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/pooler.py#L233-L263" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/layers/pooler.py", "first": 77, "last": 113, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/pooler.py#L77-L113" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/srt/layers/pooler.py", "first": 135, "last": 146, "engine": "sglang", "section": "pooling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/pooler.py#L135-L146" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "gpu_model_runner.py", "first": 3536, "last": 3536, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1907, "page": "07-architectures/07-05-lora-and-pooling-models.html", "section": "pooling", "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 3534, "last": 3553 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L3536-L3536" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/layers/base_linear.py", "first": 256, "last": 264, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/layers/base_linear.py#L256-L264" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/ops/triton_ops/kernel_utils.py", "first": 296, "last": 297, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/ops/triton_ops/kernel_utils.py#L296-L297" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "model_manager.py", "first": 334, "last": 335, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/lora/model_manager.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 1878, "page": "07-architectures/07-05-lora-and-pooling-models.html", "section": "production", "engine": "vllm", "path": "vllm/lora/model_manager.py", "first": 315, "last": 337 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/model_manager.py#L334-L335" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/layers/base_linear.py", "first": 129, "last": 150, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/layers/base_linear.py#L129-L150" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "python/sglang/kernels/ops/gemm/chunked_sgmv_shrink.py", "first": 69, "last": 79, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/gemm/chunked_sgmv_shrink.py#L69-L79" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/ops/triton_ops/lora_shrink_op.py", "first": 196, "last": 198, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/ops/triton_ops/lora_shrink_op.py#L196-L198" }, { "page": "07-architectures/07-05-lora-and-pooling-models.html", "path": "vllm/lora/punica_wrapper/punica_base.py", "first": 5, "last": 7, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/lora/punica_wrapper/punica_base.py#L5-L7" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/model_executor/models/llama.py", "first": 310, "last": 327, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L310-L327" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/model_executor/cuda_graph_buffer_registry.py", "first": 70, "last": 91, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/cuda_graph_buffer_registry.py#L70-L91" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/compilation/cuda_graph.py", "first": 161, "last": 164, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/cuda_graph.py#L161-L164" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/compilation/cuda_graph.py", "first": 346, "last": 356, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/cuda_graph.py#L346-L356" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/vllm.py", "first": 2041, "last": 2053, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L2041-L2053" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/compilation.py", "first": 692, "last": 706, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L692-L706" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/server_args.py", "first": 5179, "last": 5185, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5179-L5185" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "server_args.py", "first": 4894, "last": 4903, "engine": null, "section": "first-principles", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "chunked_prefill_size", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "chunked_prefill_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4894-L4903" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2410, "last": 2412, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2410-L2412" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/model_executor/cuda_graph_config.py", "first": 55, "last": 64, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/cuda_graph_config.py#L55-L64" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/compilation.py", "first": 53, "last": 64, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L53-L64" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/vllm.py", "first": 317, "last": 317, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L317-L317" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "cuda_graph_config.py", "first": 38, "last": 45, "engine": null, "section": "production", "declared_engine": "sglang", "declared_path": "python/sglang/srt/model_executor/cuda_graph_config.py", "reviewed_symbol": "Backend", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/model_executor/cuda_graph_config.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/model_executor/cuda_graph_config.py", "symbol": "Backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/cuda_graph_config.py#L38-L45" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/cudagraph_dispatcher.py", "first": 82, "last": 91, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/cudagraph_dispatcher.py#L82-L91" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/cudagraph_dispatcher.py", "first": 307, "last": 324, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/cudagraph_dispatcher.py#L307-L324" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/cudagraph_dispatcher.py", "first": 199, "last": 202, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/cudagraph_dispatcher.py#L199-L202" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 6959, "last": 6961, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L6959-L6961" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py", "first": 1065, "last": 1068, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py#L1065-L1068" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "runner_backend/full_cuda_graph_backend.py", "first": 105, "last": 107, "engine": null, "section": "production", "declared_engine": "sglang", "declared_path": "python/sglang/srt/model_executor/runner_backend/full_cuda_graph_backend.py", "reviewed_symbol": "warmups", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/model_executor/runner_backend/full_cuda_graph_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/model_executor/runner_backend/full_cuda_graph_backend.py", "symbol": "warmups", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/runner_backend/full_cuda_graph_backend.py#L105-L107" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/compilation.py", "first": 643, "last": 643, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L643-L643" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/vllm.py", "first": 1601, "last": 1608, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L1601-L1608" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_worker.py", "first": 726, "last": 728, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_worker.py", "reviewed_symbol": "Warmup", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_worker.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_worker.py", "symbol": "Warmup", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L726-L728" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/compilation/cuda_graph.py", "first": 286, "last": 292, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/cuda_graph.py#L286-L292" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 7041, "last": 7048, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L7041-L7048" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 559, "last": 563, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L559-L563" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 6878, "last": 6885, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L6878-L6885" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_model_runner.py", "first": 6926, "last": 6931, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "shared_memory_estimate", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "shared_memory_estimate", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L6926-L6931" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/model_executor/graph_memory_usage.py", "first": 5, "last": 12, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/graph_memory_usage.py#L5-L12" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py", "first": 191, "last": 205, "engine": "sglang", "section": "beyond-launch-overhead", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py#L191-L205" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "configs/custom_all_reduce_v2.py", "first": 129, "last": 172, "engine": null, "section": "beyond-launch-overhead", "declared_engine": "sglang", "declared_path": "python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py", "reviewed_symbol": "config", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py", "symbol": "config", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/configs/custom_all_reduce_v2.py#L129-L172" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/kernels/jit/csrc/distributed/custom_all_reduce.cuh", "first": 611, "last": 641, "engine": "sglang", "section": "beyond-launch-overhead", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/distributed/custom_all_reduce.cuh#L611-L641" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "custom_all_reduce_v2.py", "first": 13, "last": 16, "engine": null, "section": "beyond-launch-overhead", "declared_engine": "sglang", "declared_path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "reviewed_symbol": "inputs", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "symbol": "inputs", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py#L13-L16" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "custom_all_reduce.cuh", "first": 577, "last": 578, "engine": null, "section": "beyond-launch-overhead", "declared_engine": "sglang", "declared_path": "python/sglang/kernels/jit/csrc/distributed/custom_all_reduce.cuh", "reviewed_symbol": "RuntimeCheck", "resolved_engine": "sglang", "resolved_path": "python/sglang/kernels/jit/csrc/distributed/custom_all_reduce.cuh", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/kernels/jit/csrc/distributed/custom_all_reduce.cuh", "symbol": "RuntimeCheck", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/distributed/custom_all_reduce.cuh#L577-L578" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "custom_all_reduce_v2.py", "first": 253, "last": 254, "engine": null, "section": "beyond-launch-overhead", "declared_engine": "sglang", "declared_path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "reviewed_symbol": "pull_sem_mc_ptr", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "symbol": "pull_sem_mc_ptr", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py#L253-L254" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "first": 299, "last": 316, "engine": "sglang", "section": "beyond-launch-overhead", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py#L299-L316" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_model_runner.py", "first": 2252, "last": 2255, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "self.seq_lens", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "self.seq_lens", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2252-L2255" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_model_runner.py", "first": 4099, "last": 4110, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "dispatch_cudagraph", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "dispatch_cudagraph", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4099-L4110" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "cudagraph_dispatcher.py", "first": 132, "last": 156, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/cudagraph_dispatcher.py", "reviewed_symbol": "_create_padded_batch_descriptor", "resolved_engine": "vllm", "resolved_path": "vllm/v1/cudagraph_dispatcher.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/cudagraph_dispatcher.py", "symbol": "_create_padded_batch_descriptor", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/cudagraph_dispatcher.py#L132-L156" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "cuda_graph.py", "first": 360, "last": 361, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/compilation/cuda_graph.py", "reviewed_symbol": "cudagraph", "resolved_engine": "vllm", "resolved_path": "vllm/compilation/cuda_graph.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/compilation/cuda_graph.py", "symbol": "cudagraph", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/cuda_graph.py#L360-L361" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/v1/worker/workspace.py", "first": 187, "last": 193, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/workspace.py#L187-L193" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_model_runner.py", "first": 7035, "last": 7037, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "lock_workspace", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "lock_workspace", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L7035-L7037" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/compilation/monitor.py", "first": 94, "last": 99, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/monitor.py#L94-L99" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_model_runner.py", "first": 7303, "last": 7317, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "AttentionCGSupport", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "AttentionCGSupport", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L7303-L7317" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/compilation.py", "first": 1404, "last": 1412, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L1404-L1412" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/compilation.py", "first": 1463, "last": 1472, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L1463-L1472" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/server_args.py", "first": 3878, "last": 3882, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3878-L3882" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_worker.py", "first": 730, "last": 732, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_worker.py", "reviewed_symbol": "cuda_graph_memory_bytes", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_worker.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_worker.py", "symbol": "cuda_graph_memory_bytes", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L730-L732" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/server_args.py", "first": 1955, "last": 1959, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1955-L1959" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "python/sglang/srt/model_executor/runner_backend_utils/__init__.py", "first": 14, "last": 21, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/runner_backend_utils/__init__.py#L14-L21" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "gpu_model_runner.py", "first": 4149, "last": 4156, "engine": null, "section": "hands-on", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "CUDAGraphStat", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "CUDAGraphStat", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4149-L4156" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/compilation/cuda_graph.py", "first": 33, "last": 120, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/cuda_graph.py#L33-L120" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "decode_cuda_graph_runner.py", "first": 1094, "last": 1096, "engine": null, "section": "hands-on", "declared_engine": "sglang", "declared_path": "python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py", "reviewed_symbol": "capture_range", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py", "symbol": "capture_range", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/runner/decode_cuda_graph_runner.py#L1094-L1096" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "custom_all_reduce_v2.py", "first": 310, "last": 324, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "reviewed_symbol": "_pick_algo", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py", "symbol": "_pick_algo", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/custom_all_reduce_v2.py#L310-L324" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "vllm/config/compilation.py", "first": 1441, "last": 1458, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L1441-L1458" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "cudagraph_dispatcher.py", "first": 274, "last": 281, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/cudagraph_dispatcher.py", "reviewed_symbol": "keys_initialized", "resolved_engine": "vllm", "resolved_path": "vllm/v1/cudagraph_dispatcher.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/cudagraph_dispatcher.py", "symbol": "keys_initialized", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/cudagraph_dispatcher.py#L274-L281" }, { "page": "08-compilation-and-runtime/08-01-cuda-graphs.html", "path": "server_args.py", "first": 4956, "last": 4958, "engine": null, "section": "further-reading", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "15927", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "15927", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4956-L4958" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/monitor.py", "first": 17, "last": 22, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/monitor.py#L17-L22" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/model_executor/custom_op.py", "first": 296, "last": 311, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/custom_op.py#L296-L311" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/ir/ops/layernorm.py", "first": 43, "last": 62, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/ir/ops/layernorm.py#L43-L62" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 743, "last": 772, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L743-L772" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 763, "last": 779, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L763-L779" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 1134, "last": 1147, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L1134-L1147" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 577, "last": 592, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L577-L592" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "attention.py", "first": 430, "last": 430, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/attention/attention.py", "reviewed_symbol": "use_direct_call", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/attention/attention.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/attention/attention.py", "symbol": "use_direct_call", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L430-L430" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/platforms/cuda.py", "first": 577, "last": 579, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cuda.py#L577-L579" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 550, "last": 569, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L550-L569" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/partition_rules.py", "first": 14, "last": 38, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/partition_rules.py#L14-L38" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 655, "last": 684, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L655-L684" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/model_executor/models/llama.py", "first": 334, "last": 343, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L334-L343" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/decorators.py", "first": 414, "last": 446, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/decorators.py#L414-L446" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/wrapper.py", "first": 105, "last": 125, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/wrapper.py#L105-L125" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/compilation/backend.py", "first": 231, "last": 241, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/compilation/backend.py#L231-L241" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/compilation/compilation_config.py", "first": 10, "last": 16, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/compilation/compilation_config.py#L10-L16" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/layers/radix_attention.py", "first": 403, "last": 405, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/radix_attention.py#L403-L405" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/model_executor/cuda_graph_config.py", "first": 87, "last": 95, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/cuda_graph_config.py#L87-L95" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/model_executor/cuda_graph_config.py", "first": 110, "last": 119, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/cuda_graph_config.py#L110-L119" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/server_args.py", "first": 1926, "last": 1930, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1926-L1930" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/compilation/torch_compile_decoration.py", "first": 53, "last": 62, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/compilation/torch_compile_decoration.py#L53-L62" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/server_args.py", "first": 2087, "last": 2097, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2087-L2097" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/passes/pass_manager.py", "first": 155, "last": 232, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/passes/pass_manager.py#L155-L232" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/compilation/pass_manager.py", "first": 36, "last": 52, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/compilation/pass_manager.py#L36-L52" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/vllm.py", "first": 147, "last": 154, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L147-L154" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/passes/fusion/sequence_parallelism.py", "first": 202, "last": 248, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/passes/fusion/sequence_parallelism.py#L202-L248" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/passes/fusion/sequence_parallelism.py", "first": 511, "last": 521, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/passes/fusion/sequence_parallelism.py#L511-L521" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 669, "last": 687, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L669-L687" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 1058, "last": 1070, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L1058-L1070" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 780, "last": 812, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L780-L812" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 1039, "last": 1056, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L1039-L1056" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/compiler_interface.py", "first": 170, "last": 190, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/compiler_interface.py#L170-L190" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 730, "last": 775, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L730-L775" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 124, "last": 138, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L124-L138" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 580, "last": 593, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L580-L593" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/compilation/backend.py", "first": 304, "last": 343, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/compilation/backend.py#L304-L343" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/srt/compilation/cuda_piecewise_backend.py", "first": 114, "last": 126, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/compilation/cuda_piecewise_backend.py#L114-L126" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/monitor.py", "first": 61, "last": 82, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/monitor.py#L61-L82" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "docs/design/debug_vllm_compile.md", "first": 138, "last": 147, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/debug_vllm_compile.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/debug_vllm_compile.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/debug_vllm_compile.md#L138-L147" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "docs/design/debug_vllm_compile.md", "first": 154, "last": 176, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/debug_vllm_compile.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/debug_vllm_compile.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/debug_vllm_compile.md#L154-L176" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/passes/fusion/sequence_parallelism.py", "first": 592, "last": 612, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/passes/fusion/sequence_parallelism.py#L592-L612" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "docs/design/debug_vllm_compile.md", "first": 303, "last": 331, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/debug_vllm_compile.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/debug_vllm_compile.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/debug_vllm_compile.md#L303-L331" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "python/sglang/compile_deep_gemm.py", "first": 31, "last": 37, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/compile_deep_gemm.py#L31-L37" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 553, "last": 627, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L553-L627" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 1134, "last": 1185, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L1134-L1185" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 1187, "last": 1210, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L1187-L1210" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/compilation/backends.py", "first": 1030, "last": 1073, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/compilation/backends.py#L1030-L1073" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/compilation.py", "first": 780, "last": 812, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/compilation.py#L780-L812" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/model_executor/layers/activation.py", "first": 133, "last": 137, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/activation.py#L133-L137" }, { "page": "08-compilation-and-runtime/08-02-torch-compile.html", "path": "vllm/config/vllm.py", "first": 147, "last": 154, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L147-L154" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/jit/csrc/elementwise/rmsnorm.cuh", "first": 53, "last": 103, "engine": "sglang", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/csrc/elementwise/rmsnorm.cuh#L53-L103" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/ops/layernorm/norm.py", "first": 129, "last": 144, "engine": "sglang", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/layernorm/norm.py#L129-L144" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/ops/layernorm/minimax_m3_rmsnorm.py", "first": 18, "last": 43, "engine": "sglang", "section": "writing-the-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/layernorm/minimax_m3_rmsnorm.py#L18-L43" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/ops/layernorm/minimax_m3_rmsnorm.py", "first": 88, "last": 115, "engine": "sglang", "section": "writing-the-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/layernorm/minimax_m3_rmsnorm.py#L88-L115" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/ops/layernorm/minimax_m3_rmsnorm.py", "first": 46, "last": 85, "engine": "sglang", "section": "writing-the-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/layernorm/minimax_m3_rmsnorm.py#L46-L85" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/ops/attention/dsa/triton_sparse_mla.py", "first": 27, "last": 41, "engine": "sglang", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/dsa/triton_sparse_mla.py#L27-L41" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/mamba/ops/ssd_bmm.py", "first": 14, "last": 63, "engine": "vllm", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/mamba/ops/ssd_bmm.py#L14-L63" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1110, "last": 1122, "engine": "vllm", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1110-L1122" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1165, "last": 1171, "engine": "vllm", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1165-L1171" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1109, "last": 1109, "engine": "vllm", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1109-L1109" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/env_override.py", "first": 107, "last": 113, "engine": "vllm", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/env_override.py#L107-L113" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/utils/jit_monitor.py", "first": 119, "last": 141, "engine": "vllm", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/jit_monitor.py#L119-L141" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/envs.py", "first": 1177, "last": 1184, "engine": "vllm", "section": "autotuning", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1177-L1184" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 298, "last": 610, "engine": "vllm", "section": "production-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L298-L610" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 385, "last": 396, "engine": "vllm", "section": "production-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L385-L396" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/moe_align_block_size.py", "first": 11, "last": 103, "engine": "vllm", "section": "production-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/moe_align_block_size.py#L11-L103" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 404, "last": 421, "engine": "vllm", "section": "production-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L404-L421" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 519, "last": 537, "engine": "vllm", "section": "production-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L519-L537" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 577, "last": 610, "engine": "vllm", "section": "production-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L577-L610" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 836, "last": 839, "engine": "vllm", "section": "production-kernel", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L836-L839" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/srt/layers/attention/merge_state.py", "first": 12, "last": 46, "engine": "sglang", "section": "dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/merge_state.py#L12-L46" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/spec.py", "first": 29, "last": 51, "engine": "sglang", "section": "dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/spec.py#L29-L51" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/README.md", "first": 39, "last": 64, "engine": "sglang", "section": "dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/README.md#L39-L64" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/README.md", "first": 75, "last": 76, "engine": "sglang", "section": "dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/README.md#L75-L76" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm.cuh", "first": 57, "last": 84, "engine": "vllm", "section": "cutlass", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm.cuh#L57-L84" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh", "first": 102, "last": 110, "engine": null, "section": "cutlass", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh#L102-L110" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "csrc/cutlass_extensions/epilogue/broadcast_load_epilogue_c3x.hpp", "first": 33, "last": 48, "engine": "vllm", "section": "cutlass", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cutlass_extensions/epilogue/broadcast_load_epilogue_c3x.hpp#L33-L48" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "csrc/cutlass_extensions/cute_utils.cuh", "first": 13, "last": 56, "engine": "vllm", "section": "cutlass", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/cutlass_extensions/cute_utils.cuh#L13-L56" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "CMakeLists.txt", "first": 485, "last": 513, "engine": null, "section": "cutlass", "declared_engine": "vllm", "declared_path": "CMakeLists.txt", "reviewed_symbol": "CUTLASS_REVISION", "resolved_engine": "vllm", "resolved_path": "CMakeLists.txt", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "CMakeLists.txt", "symbol": "CUTLASS_REVISION", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/CMakeLists.txt#L485-L513" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "csrc/libtorch_stable/quantization/machete/Readme.md", "first": 1, "last": 3, "engine": null, "section": "cutlass", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/Readme.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/quantization/machete/Readme.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/Readme.md#L1-L3" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "csrc/libtorch_stable/quantization/machete/Readme.md", "first": 41, "last": 43, "engine": null, "section": "cutlass", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/quantization/machete/Readme.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/quantization/machete/Readme.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/machete/Readme.md#L41-L43" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 298, "last": 610, "engine": "vllm", "section": "the-map", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L298-L610" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "docs/usage/security.md", "first": 376, "last": 376, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/usage/security.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/usage/security.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/usage/security.md#L376-L376" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/srt/environ.py", "first": 1658, "last": 1658, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1658-L1658" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "fused_moe.py", "first": 417, "last": 419, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2044, "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "section": "production-kernel", "engine": "vllm", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 404, "last": 421 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L417-L419" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "fused_moe.py", "first": 847, "last": 859, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "reviewed_symbol": "use_td", "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "symbol": "use_td", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L847-L859" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/utils/jit_monitor.py", "first": 119, "last": 141, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/jit_monitor.py#L119-L141" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "test/registered/kernels/benchmark/layernorm/bench_norm.py", "first": 34, "last": 34, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "test/registered/kernels/benchmark/layernorm/bench_norm.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "test/registered/kernels/benchmark/layernorm/bench_norm.py", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/registered/kernels/benchmark/layernorm/bench_norm.py#L34-L34" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "test/registered/kernels/ops/layernorm/test_minimax_m3_rmsnorm.py", "first": 29, "last": 50, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/registered/kernels/ops/layernorm/test_minimax_m3_rmsnorm.py#L29-L50" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "python/sglang/kernels/jit/benchmark/utils.py", "first": 81, "last": 98, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/jit/benchmark/utils.py#L81-L98" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "CMakeLists.txt", "first": 486, "last": 486, "engine": null, "section": "further-reading", "declared_engine": "vllm", "declared_path": "CMakeLists.txt", "reviewed_symbol": "CUTLASS_REVISION", "resolved_engine": "vllm", "resolved_path": "CMakeLists.txt", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "CMakeLists.txt", "symbol": "CUTLASS_REVISION", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/CMakeLists.txt#L486-L486" }, { "page": "08-compilation-and-runtime/08-03-triton-and-cutlass.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1109, "last": 1109, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1109-L1109" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/weight_utils.py", "first": 957, "last": 963, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/weight_utils.py#L957-L963" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/model_loader/weight_utils.py", "first": 1107, "last": 1117, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_loader/weight_utils.py#L1107-L1117" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/weight_utils.py", "first": 849, "last": 856, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/weight_utils.py#L849-L856" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/config/load.py", "first": 62, "last": 83, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/load.py#L62-L83" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/llama.py", "first": 344, "last": 354, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L344-L354" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/llama.py", "first": 441, "last": 443, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L441-L443" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/utils.py", "first": 115, "last": 147, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/utils.py#L115-L147" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/layers/linear.py", "first": 945, "last": 960, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L945-L960" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/models/llama.py", "first": 663, "last": 678, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/llama.py#L663-L678" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/models/llama.py", "first": 714, "last": 742, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/llama.py#L714-L742" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/default_loader.py", "first": 465, "last": 470, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/default_loader.py#L465-L470" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/weight_utils.py", "first": 1243, "last": 1255, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/weight_utils.py#L1243-L1255" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/layers/linear.py", "first": 1238, "last": 1280, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L1238-L1280" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/__init__.py", "first": 32, "last": 63, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/__init__.py#L32-L63" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/model_loader/loader.py", "first": 4338, "last": 4350, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_loader/loader.py#L4338-L4350" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/__init__.py", "first": 32, "last": 63, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/__init__.py#L32-L63" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/configs/load_config.py", "first": 17, "last": 38, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/load_config.py#L17-L38" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/server_args.py", "first": 117, "last": 135, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L117-L135" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/__init__.py", "first": 48, "last": 63, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/__init__.py#L48-L63" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/base_loader.py", "first": 53, "last": 64, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/base_loader.py#L53-L64" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/default_loader.py", "first": 414, "last": 427, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/default_loader.py#L414-L427" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/llama.py", "first": 162, "last": 170, "engine": "vllm", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L162-L170" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/registry.py", "first": 153, "last": 153, "engine": "vllm", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/registry.py#L153-L153" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/models/registry.py", "first": 97, "last": 112, "engine": "sglang", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L97-L112" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/registry.py", "first": 1167, "last": 1169, "engine": "vllm", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/registry.py#L1167-L1169" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/interfaces.py", "first": 730, "last": 744, "engine": "vllm", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/interfaces.py#L730-L744" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/llama.py", "first": 444, "last": 452, "engine": "vllm", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L444-L452" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "docs/contributing/model/tests.md", "first": 10, "last": 13, "engine": "vllm", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/contributing/model/tests.md#L10-L13" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "tests/models/registry.py", "first": 382, "last": 391, "engine": "vllm", "section": "adding-a-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/models/registry.py#L382-L391" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/sharded_state_loader.py", "first": 29, "last": 38, "engine": "vllm", "section": "fast-load", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/sharded_state_loader.py#L29-L38" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/weight_cache/daemon.py", "first": 2, "last": 9, "engine": "sglang", "section": "fast-load", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/weight_cache/daemon.py#L2-L9" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/weight_cache/ipc_loader.py", "first": 51, "last": 61, "engine": "sglang", "section": "fast-load", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/weight_cache/ipc_loader.py#L51-L61" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/server_args.py", "first": 3577, "last": 3583, "engine": "sglang", "section": "fast-load", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3577-L3583" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/model_executor/model_runner_components/weight_updater.py", "first": 184, "last": 200, "engine": "sglang", "section": "fast-load", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner_components/weight_updater.py#L184-L200" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "python/sglang/srt/weight_sync/utils.py", "first": 106, "last": 121, "engine": "sglang", "section": "fast-load", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/weight_sync/utils.py#L106-L121" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/base_loader.py", "first": 36, "last": 40, "engine": "vllm", "section": "fast-load", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/base_loader.py#L36-L40" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/utils.py", "first": 411, "last": 417, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/utils.py#L411-L417" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/weight_utils.py", "first": 1365, "last": 1404, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/weight_utils.py#L1365-L1404" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/weight_utils.py", "first": 1396, "last": 1403, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/weight_utils.py#L1396-L1403" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/model_loader/sharded_state_loader.py", "first": 130, "last": 133, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/model_loader/sharded_state_loader.py#L130-L133" }, { "page": "08-compilation-and-runtime/08-04-weight-loading-and-adding-a-model.html", "path": "vllm/model_executor/models/llama.py", "first": 349, "last": 349, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/llama.py#L349-L349" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/serve/utils/api_utils.py", "first": 183, "last": 205, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/utils/api_utils.py#L183-L205" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 1199, "last": 1207, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1199-L1207" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/launchers/api_server/routers.py", "first": 39, "last": 67, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/launchers/api_server/routers.py#L39-L67" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 1209, "last": 1214, "engine": "sglang", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1209-L1214" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 258, "last": 263, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L258-L263" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/openai/completion/serving.py", "first": 117, "last": 125, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/completion/serving.py#L117-L125" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/serve/utils/tool_calls_utils.py", "first": 19, "last": 37, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/utils/tool_calls_utils.py#L19-L37" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/openai/protocol.py", "first": 1099, "last": 1122, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/protocol.py#L1099-L1122" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 607, "last": 625, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L607-L625" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/sampling/sampling_batch_info.py", "first": 134, "last": 140, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_batch_info.py#L134-L140" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/v1/sample/logits_processor/builtin.py", "first": 150, "last": 162, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/builtin.py#L150-L162" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/v1/sample/logits_processor/__init__.py", "first": 202, "last": 210, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/__init__.py#L202-L210" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/renderers/hf.py", "first": 273, "last": 292, "engine": "vllm", "section": "chat-templates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/renderers/hf.py#L273-L292" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/parser/template_manager.py", "first": 147, "last": 158, "engine": "sglang", "section": "chat-templates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/parser/template_manager.py#L147-L158" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/parser/conversation.py", "first": 1155, "last": 1159, "engine": "sglang", "section": "chat-templates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/parser/conversation.py#L1155-L1159" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/renderers/hf.py", "first": 644, "last": 672, "engine": "vllm", "section": "chat-templates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/renderers/hf.py#L644-L672" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "examples/tool_chat_template_llama3.1_json.jinja", "first": 21, "last": 34, "engine": "vllm", "section": "chat-templates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/tool_chat_template_llama3.1_json.jinja#L21-L34" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "examples/tool_chat_template_llama3.1_json.jinja", "first": 118, "last": 120, "engine": "vllm", "section": "chat-templates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/tool_chat_template_llama3.1_json.jinja#L118-L120" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 985, "last": 995, "engine": "vllm", "section": "chat-templates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L985-L995" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 899, "last": 983, "engine": "vllm", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L899-L983" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "first": 829, "last": 868, "engine": "sglang", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L829-L868" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/tool_parsers/hermes_tool_parser.py", "first": 34, "last": 40, "engine": "vllm", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tool_parsers/hermes_tool_parser.py#L34-L40" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/tool_parsers/llama_tool_parser.py", "first": 48, "last": 53, "engine": "vllm", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tool_parsers/llama_tool_parser.py#L48-L53" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/function_call/function_call_parser.py", "first": 101, "last": 111, "engine": "sglang", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/function_call/function_call_parser.py#L101-L111" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/openai/chat_completion/serving.py", "first": 905, "last": 912, "engine": "vllm", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/serving.py#L905-L912" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "first": 1764, "last": 1770, "engine": "sglang", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1764-L1770" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/serve/utils/api_utils.py", "first": 289, "last": 301, "engine": "vllm", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/utils/api_utils.py#L289-L301" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/openai/utils.py", "first": 92, "last": 106, "engine": "sglang", "section": "tools-and-streaming", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/utils.py#L92-L106" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/serve/exception_handling/error_response.py", "first": 39, "last": 59, "engine": "vllm", "section": "error-contract", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/exception_handling/error_response.py#L39-L59" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/openai/serving_base.py", "first": 110, "last": 132, "engine": "sglang", "section": "error-contract", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_base.py#L110-L132" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/serve/engine/serving.py", "first": 193, "last": 197, "engine": "vllm", "section": "error-contract", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/engine/serving.py#L193-L197" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 1875, "last": 1892, "engine": "sglang", "section": "error-contract", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L1875-L1892" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/openai/serving_base.py", "first": 40, "last": 51, "engine": "sglang", "section": "error-contract", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_base.py#L40-L51" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/config/structured_outputs.py", "first": 18, "last": 30, "engine": "vllm", "section": "error-contract", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/structured_outputs.py#L18-L30" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/serve/instrumentator/health.py", "first": 22, "last": 33, "engine": "vllm", "section": "porting-hazards", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/instrumentator/health.py#L22-L33" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 654, "last": 680, "engine": "sglang", "section": "porting-hazards", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L654-L680" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 285, "last": 287, "engine": "sglang", "section": "porting-hazards", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L285-L287" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/scale_out/render/api_router.py", "first": 26, "last": 53, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/scale_out/render/api_router.py#L26-L53" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/renderers/hf.py", "first": 644, "last": 672, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/renderers/hf.py#L644-L672" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "vllm/entrypoints/openai/chat_completion/serving.py", "first": 907, "last": 912, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/serving.py#L907-L912" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "first": 1764, "last": 1770, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1764-L1770" }, { "page": "09-serving-system/09-01-openai-api-surface.html", "path": "serving_chat.py", "first": 1112, "last": 1112, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "reviewed_symbol": "parallel_tool_calls", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "symbol": "parallel_tool_calls", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1112-L1112" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 23, "last": 28, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L23-L28" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 235, "last": 238, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L235-L238" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/detokenizer_utils.py", "first": 260, "last": 264, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/detokenizer_utils.py#L260-L264" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "tests/tokenizers_/test_detokenize.py", "first": 108, "last": 108, "engine": null, "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/tokenizers_/test_detokenize.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/tokenizers_/test_detokenize.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/tokenizers_/test_detokenize.py#L108-L108" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "tests/tokenizers_/test_detokenize.py", "first": 105, "last": 107, "engine": null, "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/tokenizers_/test_detokenize.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/tokenizers_/test_detokenize.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/tokenizers_/test_detokenize.py#L105-L107" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/detokenizer_utils.py", "first": 195, "last": 197, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/detokenizer_utils.py#L195-L197" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/detokenizer_utils.py", "first": 57, "last": 59, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/detokenizer_utils.py#L57-L59" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 144, "last": 144, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L144-L144" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1427, "last": 1427, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1427-L1427" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/detokenizer_utils.py", "first": 238, "last": 268, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/detokenizer_utils.py#L238-L268" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/detokenizer_utils.py", "first": 129, "last": 140, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/detokenizer_utils.py#L129-L140" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 374, "last": 395, "engine": "sglang", "section": "partial-utf8", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L374-L395" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/utils.py", "first": 354, "last": 370, "engine": "sglang", "section": "partial-utf8", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/utils.py#L354-L370" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/async_llm.py", "first": 140, "last": 145, "engine": "vllm", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L140-L145" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/async_llm.py", "first": 697, "last": 707, "engine": "vllm", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L697-L707" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/envs.py", "first": 1430, "last": 1432, "engine": "vllm", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1430-L1432" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 55, "last": 66, "engine": "vllm", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L55-L66" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 180, "last": 187, "engine": "vllm", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L180-L187" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 239, "last": 247, "engine": "vllm", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L239-L247" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 516, "last": 533, "engine": "sglang", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L516-L533" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 251, "last": 267, "engine": "sglang", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L251-L267" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 325, "last": 335, "engine": "sglang", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L325-L335" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 238, "last": 241, "engine": "sglang", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L238-L241" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 336, "last": 337, "engine": "sglang", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L336-L337" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 363, "last": 373, "engine": "sglang", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L363-L373" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 57, "last": 61, "engine": "sglang", "section": "where-it-runs", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L57-L61" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/sampling_params.py", "first": 646, "last": 650, "engine": "vllm", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L646-L650" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/sampling/sampling_params.py", "first": 318, "last": 322, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_params.py#L318-L322" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 85, "last": 91, "engine": "vllm", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L85-L91" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 149, "last": 165, "engine": "vllm", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L149-L165" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 331, "last": 349, "engine": "vllm", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L331-L349" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 326, "last": 329, "engine": "vllm", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L326-L329" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1459, "last": 1469, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1459-L1469" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/sampling/sampling_params.py", "first": 221, "last": 228, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_params.py#L221-L228" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "first": 381, "last": 384, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_streamer.py#L381-L384" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1488, "last": 1492, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1488-L1492" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 189, "last": 195, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L189-L195" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/sampling/sampling_params.py", "first": 78, "last": 78, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_params.py#L78-L78" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 187, "last": 187, "engine": "sglang", "section": "stop-strings", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L187-L187" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/utils/hf_transformers/tokenizer.py", "first": 432, "last": 435, "engine": "sglang", "section": "cost", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/hf_transformers/tokenizer.py#L432-L435" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/config/scheduler.py", "first": 153, "last": 157, "engine": "vllm", "section": "cost", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L153-L157" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/server_args.py", "first": 1480, "last": 1483, "engine": "sglang", "section": "cost", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1480-L1483" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/utils/hf_transformers/tokenizer.py", "first": 489, "last": 498, "engine": "sglang", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/hf_transformers/tokenizer.py#L489-L498" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/hf.py", "first": 107, "last": 111, "engine": "vllm", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/hf.py#L107-L111" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/entrypoints/serve/tokenize/protocol.py", "first": 78, "last": 87, "engine": "vllm", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/tokenize/protocol.py#L78-L87" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/entrypoints/serve/tokenize/protocol.py", "first": 28, "last": 33, "engine": "vllm", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/tokenize/protocol.py#L28-L33" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/tokenizer/tiktoken_tokenizer.py", "first": 132, "last": 132, "engine": "sglang", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/tokenizer/tiktoken_tokenizer.py#L132-L132" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/detokenizer_utils.py", "first": 66, "last": 76, "engine": "vllm", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/detokenizer_utils.py#L66-L76" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/async_dynamic_batch_tokenizer.py", "first": 77, "last": 101, "engine": "sglang", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/async_dynamic_batch_tokenizer.py#L77-L101" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/async_dynamic_batch_tokenizer.py", "first": 43, "last": 44, "engine": "sglang", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/async_dynamic_batch_tokenizer.py#L43-L44" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/async_dynamic_batch_tokenizer.py", "first": 143, "last": 147, "engine": "sglang", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/async_dynamic_batch_tokenizer.py#L143-L147" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/server_args.py", "first": 3466, "last": 3480, "engine": "sglang", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3466-L3480" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/hf.py", "first": 25, "last": 57, "engine": "vllm", "section": "tokenizing-in", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/hf.py#L25-L57" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/async_llm.py", "first": 685, "last": 691, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L685-L691" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/output_processor.py", "first": 603, "last": 603, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L603-L603" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/output_processor.py", "first": 676, "last": 682, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L676-L682" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 96, "last": 143, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L96-L143" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 211, "last": 222, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L211-L222" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/output_processor.py", "first": 407, "last": 407, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L407-L407" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/output_processor.py", "first": 716, "last": 722, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L716-L722" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1552, "last": 1560, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1552-L1560" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "first": 412, "last": 423, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_streamer.py#L412-L423" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/tokenizers/detokenizer_utils.py", "first": 260, "last": 268, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tokenizers/detokenizer_utils.py#L260-L268" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 374, "last": 395, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L374-L395" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "tests/tokenizers_/test_detokenize.py", "first": 104, "last": 110, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/tokenizers_/test_detokenize.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/tokenizers_/test_detokenize.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/tokenizers_/test_detokenize.py#L104-L110" }, { "page": "09-serving-system/09-02-tokenization-and-detokenization.html", "path": "vllm/v1/engine/detokenizer.py", "first": 121, "last": 121, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/detokenizer.py#L121-L121" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 1616, "last": 1624, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1616-L1624" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/launchers/api_server/routers.py", "first": 12, "last": 20, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/launchers/api_server/routers.py#L12-L20" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/generate/api_router.py", "first": 21, "last": 26, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/generate/api_router.py#L21-L26" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/openai/chat_completion/api_router.py", "first": 40, "last": 53, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/api_router.py#L40-L53" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/renderers/online_renderer.py", "first": 117, "last": 218, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/renderers/online_renderer.py#L117-L218" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/renderers/base.py", "first": 1077, "last": 1091, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/renderers/base.py#L1077-L1091" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/renderers/base.py", "first": 85, "last": 99, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/renderers/base.py#L85-L99" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/config/model.py", "first": 367, "last": 367, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L367-L367" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 737, "last": 739, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L737-L739" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/openai/chat_completion/serving.py", "first": 370, "last": 398, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/serving.py#L370-L398" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/async_llm.py", "first": 424, "last": 436, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L424-L436" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/__init__.py", "first": 107, "last": 122, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/__init__.py#L107-L122" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core_client.py", "first": 1108, "last": 1151, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L1108-L1151" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core_client.py", "first": 554, "last": 563, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L554-L563" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core.py", "first": 1690, "last": 1697, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1690-L1697" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/utils/network_utils.py", "first": 336, "last": 342, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/network_utils.py#L336-L342" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core.py", "first": 1674, "last": 1775, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1674-L1775" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core.py", "first": 1391, "last": 1402, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1391-L1402" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core.py", "first": 478, "last": 478, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L478-L478" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core.py", "first": 590, "last": 610, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L590-L610" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 639, "last": 639, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L639-L639" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2030, "last": 2040, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2030-L2040" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/core/sched/utils.py", "first": 94, "last": 130, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/utils.py#L94-L130" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core.py", "first": 1812, "last": 1826, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1812-L1826" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core_client.py", "first": 1041, "last": 1046, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L1041-L1046" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/async_llm.py", "first": 686, "last": 729, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L686-L729" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/envs.py", "first": 169, "last": 169, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L169-L169" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/output_processor.py", "first": 676, "last": 702, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L676-L702" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/output_processor.py", "first": 48, "last": 75, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L48-L75" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/async_llm.py", "first": 601, "last": 614, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L601-L614" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/openai/chat_completion/serving.py", "first": 645, "last": 816, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/serving.py#L645-L816" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/openai/chat_completion/api_router.py", "first": 74, "last": 74, "engine": "vllm", "section": "vllm-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/api_router.py#L74-L74" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "serving.py", "first": 912, "last": 912, "engine": null, "section": "vllm-trace", "declared_engine": "vllm", "declared_path": "vllm/entrypoints/openai/chat_completion/serving.py", "reviewed_symbol": "yield \"data: [DONE]\\n\\n\"", "resolved_engine": "vllm", "resolved_path": "vllm/entrypoints/openai/chat_completion/serving.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/entrypoints/openai/chat_completion/serving.py", "symbol": "yield \"data: [DONE]\\n\\n\"", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/serving.py#L912-L912" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 1722, "last": 1729, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L1722-L1729" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/entrypoints/openai/serving_base.py", "first": 73, "last": 109, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_base.py#L73-L109" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "first": 1348, "last": 1359, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1348-L1359" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "first": 1485, "last": 1504, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1485-L1504" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/openai/chat_completion/serving.py", "first": 236, "last": 240, "engine": "vllm", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/serving.py#L236-L240" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 810, "last": 818, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L810-L818" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/io_struct.py", "first": 2421, "last": 2426, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L2421-L2426" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/io_struct.py", "first": 2381, "last": 2381, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L2381-L2381" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/environ.py", "first": 337, "last": 337, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L337-L337" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "io_struct.py", "first": 106, "last": 116, "engine": null, "section": "sglang-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/io_struct.py", "reviewed_symbol": "PickleWrapper", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/io_struct.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/io_struct.py", "symbol": "PickleWrapper", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L106-L116" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 553, "last": 555, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L553-L555" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/utils/network.py", "first": 248, "last": 253, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/network.py#L248-L253" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler_components/request_receiver.py", "first": 118, "last": 125, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/request_receiver.py#L118-L125" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1783, "last": 1800, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1783-L1800" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "scheduler.py", "first": 1923, "last": 1923, "engine": null, "section": "sglang-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/scheduler.py", "reviewed_symbol": "_request_dispatcher", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/scheduler.py", "symbol": "_request_dispatcher", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1923-L1923" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "scheduler.py", "first": 2435, "last": 2470, "engine": null, "section": "sglang-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/scheduler.py", "reviewed_symbol": "req = Req", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/scheduler.py", "symbol": "req = Req", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L2435-L2470" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 2780, "last": 2788, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L2780-L2788" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler_components/batch_result_processor.py", "first": 862, "last": 869, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/batch_result_processor.py#L862-L869" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1632, "last": 1672, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1632-L1672" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "first": 362, "last": 384, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_streamer.py#L362-L384" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/server_args.py", "first": 1480, "last": 1484, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1480-L1484" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/output_processor.py", "first": 299, "last": 313, "engine": "vllm", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L299-L313" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "output_streamer.py", "first": 173, "last": 182, "engine": null, "section": "sglang-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "reviewed_symbol": "detokenizer", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "symbol": "detokenizer", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_streamer.py#L173-L182" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 167, "last": 175, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L167-L175" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "tokenizer_manager.py", "first": 2151, "last": 2164, "engine": null, "section": "sglang-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/tokenizer_manager.py", "reviewed_symbol": "handle_loop", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/tokenizer_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/tokenizer_manager.py", "symbol": "handle_loop", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2151-L2164" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 2438, "last": 2446, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2438-L2446" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/server_args.py", "first": 1485, "last": 1489, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1485-L1489" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 1713, "last": 1727, "engine": "sglang", "section": "sglang-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1713-L1727" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "serving_chat.py", "first": 1544, "last": 1546, "engine": null, "section": "sglang-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "reviewed_symbol": "content", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "symbol": "content", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1544-L1546" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "serving_chat.py", "first": 1770, "last": 1770, "engine": null, "section": "sglang-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "reviewed_symbol": "yield \"data: [DONE]\\n\\n\"", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "symbol": "yield \"data: [DONE]\\n\\n\"", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1770-L1770" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "serving_base.py", "first": 79, "last": 79, "engine": null, "section": "latency", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/entrypoints/openai/serving_base.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2253, "page": "09-serving-system/09-03-request-lifecycle.html", "section": "sglang-trace", "engine": "sglang", "path": "python/sglang/srt/entrypoints/openai/serving_base.py", "first": 73, "last": 109 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_base.py#L79-L79" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/renderers/base.py", "first": 1075, "last": 1075, "engine": "vllm", "section": "latency", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/renderers/base.py#L1075-L1075" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "tokenizer_manager.py", "first": 1742, "last": 1747, "engine": null, "section": "latency", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/tokenizer_manager.py", "reviewed_symbol": "set_response_sent_to_client_time", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/tokenizer_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/tokenizer_manager.py", "symbol": "set_response_sent_to_client_time", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1742-L1747" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/entrypoints/serve/utils/api_utils.py", "first": 36, "last": 87, "engine": "vllm", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/utils/api_utils.py#L36-L87" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/async_llm.py", "first": 616, "last": 758, "engine": "vllm", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L616-L758" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "output_processor.py", "first": 508, "last": 513, "engine": null, "section": "abort", "declared_engine": "vllm", "declared_path": "vllm/v1/engine/output_processor.py", "reviewed_symbol": "request_ids_to_abort", "resolved_engine": "vllm", "resolved_path": "vllm/v1/engine/output_processor.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/engine/output_processor.py", "symbol": "request_ids_to_abort", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L508-L513" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/engine/core.py", "first": 1767, "last": 1775, "engine": "vllm", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1767-L1775" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "core.py", "first": 607, "last": 607, "engine": null, "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/engine/core.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2238, "page": "09-serving-system/09-03-request-lifecycle.html", "section": "vllm-trace", "engine": "vllm", "path": "vllm/v1/engine/core.py", "first": 590, "last": 610 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L607-L607" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2400, "last": 2454, "engine": "vllm", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2400-L2454" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 2112, "last": 2120, "engine": "sglang", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2112-L2120" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "tokenizer_manager.py", "first": 1948, "last": 1953, "engine": null, "section": "abort", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/tokenizer_manager.py", "reviewed_symbol": "rid_to_state", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/tokenizer_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/tokenizer_manager.py", "symbol": "rid_to_state", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1948-L1953" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/environ.py", "first": 1217, "last": 1217, "engine": "sglang", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1217-L1217" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 4554, "last": 4691, "engine": "sglang", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L4554-L4691" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/mem_cache/common.py", "first": 214, "last": 219, "engine": "sglang", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/common.py#L214-L219" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 459, "last": 461, "engine": "sglang", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L459-L461" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2463, "last": 2475, "engine": "vllm", "section": "abort", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2463-L2475" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "tokenizer_manager.py", "first": 822, "last": 831, "engine": null, "section": "abort", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/tokenizer_manager.py", "reviewed_symbol": "_discard_pending_req_states", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/tokenizer_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/tokenizer_manager.py", "symbol": "_discard_pending_req_states", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L822-L831" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/utils/network_utils.py", "first": 336, "last": 342, "engine": "vllm", "section": "backpressure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/network_utils.py#L336-L342" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/utils/network.py", "first": 248, "last": 253, "engine": "sglang", "section": "backpressure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/network.py#L248-L253" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/server_args.py", "first": 792, "last": 796, "engine": "sglang", "section": "backpressure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L792-L796" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 2829, "last": 2874, "engine": "sglang", "section": "backpressure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L2829-L2874" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "scheduler.py", "first": 2878, "last": 2905, "engine": null, "section": "backpressure", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/scheduler.py", "reviewed_symbol": "_abort_on_waiting_timeout", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/scheduler.py", "symbol": "_abort_on_waiting_timeout", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L2878-L2905" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "tokenizer_manager.py", "first": 1644, "last": 1683, "engine": null, "section": "backpressure", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/tokenizer_manager.py", "reviewed_symbol": "_handle_abort_finish_reason", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/tokenizer_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/tokenizer_manager.py", "symbol": "_handle_abort_finish_reason", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1644-L1683" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "serving_chat.py", "first": 1590, "last": 1605, "engine": null, "section": "backpressure", "declared_engine": "sglang", "declared_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "reviewed_symbol": "explicit", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/entrypoints/openai/serving_chat.py", "symbol": "explicit", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_chat.py#L1590-L1605" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/environ.py", "first": 554, "last": 554, "engine": "sglang", "section": "backpressure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L554-L554" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 2878, "last": 2880, "engine": "sglang", "section": "backpressure", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L2878-L2880" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "serving.py", "first": 647, "last": 653, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/entrypoints/openai/chat_completion/serving.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2249, "page": "09-serving-system/09-03-request-lifecycle.html", "section": "vllm-trace", "engine": "vllm", "path": "vllm/entrypoints/openai/chat_completion/serving.py", "first": 645, "last": 816 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/serving.py#L647-L653" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "async_llm.py", "first": 725, "last": 729, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/engine/async_llm.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2244, "page": "09-serving-system/09-03-request-lifecycle.html", "section": "vllm-trace", "engine": "vllm", "path": "vllm/v1/engine/async_llm.py", "first": 686, "last": 729 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L725-L729" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "tokenizer_manager.py", "first": 3128, "last": 3132, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/tokenizer_manager.py", "reviewed_symbol": "rid_to_state", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/tokenizer_manager.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/tokenizer_manager.py", "symbol": "rid_to_state", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L3128-L3132" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 143, "last": 143, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L143-L143" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 363, "last": 373, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L363-L373" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "vllm/config/model.py", "first": 367, "last": 367, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L367-L367" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "output_processor.py", "first": 299, "last": 313, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/engine/output_processor.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2273, "page": "09-serving-system/09-03-request-lifecycle.html", "section": "sglang-trace", "engine": "vllm", "path": "vllm/v1/engine/output_processor.py", "first": 299, "last": 313 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L299-L313" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "output_streamer.py", "first": 370, "last": 380, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2271, "page": "09-serving-system/09-03-request-lifecycle.html", "section": "sglang-trace", "engine": "sglang", "path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "first": 362, "last": 384 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_streamer.py#L370-L380" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "chat_completion/protocol.py", "first": 737, "last": 739, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/entrypoints/openai/chat_completion/protocol.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2227, "page": "09-serving-system/09-03-request-lifecycle.html", "section": "vllm-trace", "engine": "vllm", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 737, "last": 739 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L737-L739" }, { "page": "09-serving-system/09-03-request-lifecycle.html", "path": "output_processor.py", "first": 292, "last": 297, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/engine/output_processor.py", "reviewed_symbol": "finished", "resolved_engine": "vllm", "resolved_path": "vllm/v1/engine/output_processor.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/engine/output_processor.py", "symbol": "finished", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L292-L297" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "docs/deployment/integrations/llm-d.md", "first": 9, "last": 11, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/deployment/integrations/llm-d.md#L9-L11" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/factory.rs", "first": 76, "last": 91, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/factory.rs#L76-L91" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 21, "last": 32, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L21-L32" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 405, "last": 424, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L405-L424" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 436, "last": 450, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L436-L450" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 469, "last": 471, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L469-L471" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 127, "last": 147, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L127-L147" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/tree.rs", "first": 718, "last": 753, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/tree.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/policies/tree.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/tree.rs#L718-L753" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/routers/http/pd_router.rs", "first": 950, "last": 960, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/routers/http/pd_router.rs#L950-L960" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/consistent_hashing.rs", "first": 136, "last": 150, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/consistent_hashing.rs#L136-L150" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/manual.rs", "first": 1, "last": 14, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/manual.rs#L1-L14" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "manual.rs", "first": 170, "last": 192, "engine": null, "section": "production", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/policies/manual.rs", "reviewed_symbol": "routing_map", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/manual.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/policies/manual.rs", "symbol": "routing_map", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/manual.rs#L170-L192" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py", "first": 336, "last": 341, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py#L336-L341" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py", "first": 250, "last": 268, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/disaggregated/disaggregated_serving/disagg_proxy_demo.py#L250-L268" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "vllm/distributed/kv_events.py", "first": 50, "last": 101, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_events.py#L50-L101" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "vllm/distributed/kv_events.py", "first": 290, "last": 313, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_events.py#L290-L313" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "vllm/config/kv_events.py", "first": 27, "last": 36, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/kv_events.py#L27-L36" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "python/sglang/srt/disaggregation/kv_events.py", "first": 112, "last": 135, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/kv_events.py#L112-L135" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "docs/deployment/integrations/llm-d.md", "first": 23, "last": 23, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/deployment/integrations/llm-d.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/deployment/integrations/llm-d.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/deployment/integrations/llm-d.md#L23-L23" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "first": 54, "last": 70, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L54-L70" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "mini_lb.py", "first": 103, "last": 112, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "reviewed_symbol": "select_pair", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "symbol": "select_pair", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L103-L112" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 76, "last": 91, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/cache_aware.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/policies/cache_aware.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L76-L91" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "first": 356, "last": 383, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L356-L383" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "mini_lb.py", "first": 436, "last": 437, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "reviewed_symbol": "_generate_bootstrap_room", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "symbol": "_generate_bootstrap_room", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L436-L437" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "python/sglang/srt/managers/disagg_service.py", "first": 21, "last": 30, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/disagg_service.py#L21-L30" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "first": 129, "last": 141, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L129-L141" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/routers/http/pd_router.rs", "first": 711, "last": 732, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/routers/http/pd_router.rs#L711-L732" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "mini_lb.py", "first": 143, "last": 156, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "reviewed_symbol": "return_logprob", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "symbol": "return_logprob", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L143-L156" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/config/types.rs", "first": 512, "last": 515, "engine": "sglang", "section": "autoscaling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/config/types.rs#L512-L515" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/service_discovery.rs", "first": 144, "last": 150, "engine": "sglang", "section": "autoscaling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/service_discovery.rs#L144-L150" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/core/worker.rs", "first": 568, "last": 579, "engine": "sglang", "section": "autoscaling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/core/worker.rs#L568-L579" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "worker.rs", "first": 733, "last": 782, "engine": null, "section": "autoscaling", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/core/worker.rs", "reviewed_symbol": "check_health_async", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/core/worker.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/core/worker.rs", "symbol": "check_health_async", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/core/worker.rs#L733-L782" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "mini_lb.py", "first": 248, "last": 257, "engine": null, "section": "autoscaling", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "reviewed_symbol": "health_generate", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "symbol": "health_generate", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L248-L257" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/core/worker_registry.rs", "first": 364, "last": 370, "engine": null, "section": "autoscaling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/core/worker_registry.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/core/worker_registry.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/core/worker_registry.rs#L364-L370" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/server.rs", "first": 1061, "last": 1065, "engine": null, "section": "autoscaling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/server.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/server.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/server.rs#L1061-L1065" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/main.rs", "first": 290, "last": 292, "engine": null, "section": "autoscaling", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/main.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/main.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/main.rs#L290-L292" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/mod.rs", "first": 96, "last": 116, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/mod.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/policies/mod.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/mod.rs#L96-L116" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/router_args.py", "first": 53, "last": 64, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/router_args.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/bindings/python/src/sglang_router/router_args.py", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/router_args.py#L53-L64" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/core/worker_manager.rs", "first": 350, "last": 357, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/core/worker_manager.rs#L350-L357" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "cache_aware.rs", "first": 492, "last": 512, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/policies/cache_aware.rs", "reviewed_symbol": "Selected", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/cache_aware.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "symbol": "Selected", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L492-L512" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 514, "last": 521, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L514-L521" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "pd_router.rs", "first": 728, "last": 732, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/routers/http/pd_router.rs", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2344, "page": "09-serving-system/09-04-routers-and-load-balancing.html", "section": "worked-trace", "engine": "sglang", "path": "sgl-model-gateway/src/routers/http/pd_router.rs", "first": 711, "last": 732 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/routers/http/pd_router.rs#L728-L732" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "python/sglang/srt/disaggregation/kv_events.py", "first": 112, "last": 135, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/kv_events.py#L112-L135" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "python/sglang/srt/managers/scheduler_components/kv_events_publisher.py", "first": 45, "last": 75, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/kv_events_publisher.py#L45-L75" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "python/sglang/srt/server_args.py", "first": 1650, "last": 1654, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1650-L1654" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "examples/disaggregated/disaggregated_serving/kv_events.sh", "first": 37, "last": 45, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "examples/disaggregated/disaggregated_serving/kv_events.sh", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "examples/disaggregated/disaggregated_serving/kv_events.sh", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/disaggregated/disaggregated_serving/kv_events.sh#L37-L45" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "cache_aware.rs", "first": 354, "last": 354, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/policies/cache_aware.rs", "reviewed_symbol": "insert", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/cache_aware.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "symbol": "insert", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L354-L354" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "cache_aware.rs", "first": 436, "last": 466, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/policies/cache_aware.rs", "reviewed_symbol": "result", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/cache_aware.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "symbol": "result", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L436-L466" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "vllm/config/kv_events.py", "first": 37, "last": 40, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/kv_events.py#L37-L40" }, { "page": "09-serving-system/09-04-routers-and-load-balancing.html", "path": "pd_router.rs", "first": 697, "last": 699, "engine": null, "section": "further-reading", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/routers/http/pd_router.rs", "reviewed_symbol": "handler", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/routers/http/pd_router.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/routers/http/pd_router.rs", "symbol": "handler", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/routers/http/pd_router.rs#L697-L699" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/metrics/loggers.py", "first": 829, "last": 854, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L829-L854" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 1671, "last": 1680, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L1671-L1680" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "loggers.py", "first": 796, "last": 804, "engine": null, "section": "first-principles", "declared_engine": "vllm", "declared_path": "vllm/v1/metrics/loggers.py", "reviewed_symbol": "histogram_time_to_first_token", "resolved_engine": "vllm", "resolved_path": "vllm/v1/metrics/loggers.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/metrics/loggers.py", "symbol": "histogram_time_to_first_token", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L796-L804" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "metrics_collector.py", "first": 1623, "last": 1644, "engine": null, "section": "first-principles", "declared_engine": "sglang", "declared_path": "python/sglang/srt/observability/metrics_collector.py", "reviewed_symbol": "bucket_time_to_first_token", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/observability/metrics_collector.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/observability/metrics_collector.py", "symbol": "bucket_time_to_first_token", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L1623-L1644" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/server_args.py", "first": 1604, "last": 1618, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1604-L1618" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/metrics/loggers.py", "first": 74, "last": 88, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L74-L88" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 577, "last": 580, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L577-L580" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/metrics/loggers.py", "first": 256, "last": 298, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L256-L298" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/envs.py", "first": 47, "last": 47, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L47-L47" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/managers/scheduler_components/metrics_reporter.py", "first": 811, "last": 811, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/metrics_reporter.py#L811-L811" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/metrics/stats.py", "first": 516, "last": 526, "engine": "vllm", "section": "saturation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L516-L526" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3619, "last": 3631, "engine": "sglang", "section": "saturation", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3619-L3631" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "loggers.py", "first": 513, "last": 523, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/metrics/loggers.py", "reviewed_symbol": "gauge_waiting_by_reason", "resolved_engine": "vllm", "resolved_path": "vllm/v1/metrics/loggers.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/metrics/loggers.py", "symbol": "gauge_waiting_by_reason", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L513-L523" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 778, "last": 809, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L778-L809" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 550, "last": 558, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L550-L558" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/engine/async_llm.py", "first": 940, "last": 943, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L940-L943" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/utils/rank_consensus_checker.py", "first": 405, "last": 430, "engine": "sglang", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/rank_consensus_checker.py#L405-L430" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "docs/usage/troubleshooting.md", "first": 33, "last": 41, "engine": null, "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/usage/troubleshooting.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/usage/troubleshooting.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/usage/troubleshooting.md#L33-L41" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/envs.py", "first": 243, "last": 243, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L243-L243" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/fault_tolerance/engine_core_sentinel.py", "first": 80, "last": 106, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/fault_tolerance/engine_core_sentinel.py#L80-L106" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/fault_tolerance/engine_core_sentinel.py", "first": 122, "last": 134, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/fault_tolerance/engine_core_sentinel.py#L122-L134" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 305, "last": 324, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L305-L324" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/engine/exceptions.py", "first": 12, "last": 13, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/exceptions.py#L12-L13" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/utils/watchdog.py", "first": 133, "last": 163, "engine": "sglang", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/watchdog.py#L133-L163" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/server_args.py", "first": 1261, "last": 1264, "engine": "sglang", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1261-L1264" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/utils/watchdog.py", "first": 166, "last": 175, "engine": "sglang", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/watchdog.py#L166-L175" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/envs.py", "first": 247, "last": 247, "engine": "vllm", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L247-L247" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "loggers.py", "first": 572, "last": 583, "engine": null, "section": "failure-taxonomy", "declared_engine": "vllm", "declared_path": "vllm/v1/metrics/loggers.py", "reviewed_symbol": "counter_corrupted_requests", "resolved_engine": "vllm", "resolved_path": "vllm/v1/metrics/loggers.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/metrics/loggers.py", "symbol": "counter_corrupted_requests", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L572-L583" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1655, "last": 1669, "engine": "sglang", "section": "failure-taxonomy", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1655-L1669" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/v1/engine/output_processor.py", "first": 805, "last": 811, "engine": "vllm", "section": "tracing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L805-L811" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/tracing/utils.py", "first": 24, "last": 45, "engine": "vllm", "section": "tracing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tracing/utils.py#L24-L45" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/tracing/utils.py", "first": 11, "last": 12, "engine": "vllm", "section": "tracing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/tracing/utils.py#L11-L12" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/observability/req_time_stats.py", "first": 105, "last": 131, "engine": "sglang", "section": "tracing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/req_time_stats.py#L105-L131" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "metrics_collector.py", "first": 733, "last": 739, "engine": null, "section": "tracing", "declared_engine": "sglang", "declared_path": "python/sglang/srt/observability/metrics_collector.py", "reviewed_symbol": "per_stage_req_latency_seconds", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/observability/metrics_collector.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/observability/metrics_collector.py", "symbol": "per_stage_req_latency_seconds", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L733-L739" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/observability/trace.py", "first": 36, "last": 37, "engine": "sglang", "section": "tracing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/trace.py#L36-L37" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/observability/trace.py", "first": 133, "last": 142, "engine": "sglang", "section": "tracing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/trace.py#L133-L142" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/engine/arg_utils.py", "first": 2879, "last": 2879, "engine": "vllm", "section": "what-to-log", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2879-L2879" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/server_args.py", "first": 1518, "last": 1530, "engine": "sglang", "section": "what-to-log", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1518-L1530" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/utils/request_logger.py", "first": 30, "last": 41, "engine": "sglang", "section": "what-to-log", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/request_logger.py#L30-L41" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 2956, "last": 2995, "engine": "sglang", "section": "what-to-log", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2956-L2995" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/entrypoints/serve/instrumentator/health.py", "first": 22, "last": 33, "engine": "vllm", "section": "recovery", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/instrumentator/health.py#L22-L33" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "vllm/entrypoints/serve/instrumentator/basic.py", "first": 30, "last": 49, "engine": "vllm", "section": "recovery", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/instrumentator/basic.py#L30-L49" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 654, "last": 675, "engine": "sglang", "section": "recovery", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L654-L675" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 3567, "last": 3583, "engine": "sglang", "section": "recovery", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L3567-L3583" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "loggers.py", "first": 859, "last": 882, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/metrics/loggers.py", "reviewed_symbol": "histogram_request_time_per_output_token", "resolved_engine": "vllm", "resolved_path": "vllm/v1/metrics/loggers.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/metrics/loggers.py", "symbol": "histogram_request_time_per_output_token", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L859-L882" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "multiproc_executor.py", "first": 305, "last": 305, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/executor/multiproc_executor.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2389, "page": "09-serving-system/09-05-observability-and-failure.html", "section": "failure-taxonomy", "engine": "vllm", "path": "vllm/v1/executor/multiproc_executor.py", "first": 305, "last": 324 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L305-L305" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "async_llm.py", "first": 305, "last": 305, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/engine/async_llm.py", "reviewed_symbol": "EngineDeadError", "resolved_engine": "vllm", "resolved_path": "vllm/v1/engine/async_llm.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/engine/async_llm.py", "symbol": "EngineDeadError", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L305-L305" }, { "page": "09-serving-system/09-05-observability-and-failure.html", "path": "stats.py", "first": 437, "last": 437, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/metrics/stats.py", "reviewed_symbol": "inter_token_latencies_iter", "resolved_engine": "vllm", "resolved_path": "vllm/v1/metrics/stats.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/metrics/stats.py", "symbol": "inter_token_latencies_iter", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L437-L437" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/serve.py", "first": 456, "last": 467, "engine": "vllm", "section": "four-axes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L456-L467" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/benchmark/serving.py", "first": 1083, "last": 1092, "engine": "sglang", "section": "four-axes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1083-L1092" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/serve.py", "first": 961, "last": 972, "engine": "vllm", "section": "four-axes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L961-L972" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 1615, "last": 1635, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L1615-L1635" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/benchmark/datasets/__init__.py", "first": 18, "last": 32, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/__init__.py#L18-L32" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/datasets/utils.py", "first": 74, "last": 99, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/utils.py#L74-L99" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/benchmark/datasets/common.py", "first": 56, "last": 64, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/common.py#L56-L64" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/benchmark/datasets/sharegpt.py", "first": 88, "last": 88, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/sharegpt.py#L88-L88" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 1393, "last": 1396, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L1393-L1396" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 341, "last": 356, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L341-L356" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/v1/metrics/stats.py", "first": 239, "last": 255, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L239-L255" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/v1/metrics/stats.py", "first": 258, "last": 276, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L258-L276" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/v1/metrics/loggers.py", "first": 730, "last": 747, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L730-L747" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/srt/observability/request_metrics_exporter.py", "first": 33, "last": 62, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/request_metrics_exporter.py#L33-L62" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 2238, "last": 2245, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2238-L2245" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 290, "last": 294, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L290-L294" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/datasets/utils.py", "first": 64, "last": 101, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/utils.py#L64-L101" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/benchmark/datasets/common.py", "first": 56, "last": 64, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/common.py#L56-L64" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 1502, "last": 1533, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L1502-L1533" }, { "page": "10-benchmarking/10-01-workload-characterization.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 290, "last": 294, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L290-L294" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/bench_serving.py", "first": 8, "last": 21, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/bench_serving.py#L8-L21" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 462, "last": 485, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L462-L485" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 497, "last": 504, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L497-L504" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 1080, "last": 1092, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1080-L1092" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 805, "last": 822, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L805-L822" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 71, "last": 82, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L71-L82" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 269, "last": 269, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L269-L269" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 224, "last": 247, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L224-L247" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 333, "last": 348, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L333-L348" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 749, "last": 760, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L749-L760" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 1116, "last": 1146, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1116-L1146" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 1581, "last": 1600, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1581-L1600" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "endpoint_request_func.py", "first": 238, "last": 240, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/benchmarks/lib/endpoint_request_func.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 2443, "page": "10-benchmarking/10-02-benchmark-harnesses.html", "section": "production", "engine": "vllm", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 224, "last": 247 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L238-L240" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "endpoint_request_func.py", "first": 256, "last": 257, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/benchmarks/lib/endpoint_request_func.py", "reviewed_symbol": "output", "resolved_engine": "vllm", "resolved_path": "vllm/benchmarks/lib/endpoint_request_func.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "symbol": "output", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L256-L257" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "serve.py", "first": 2165, "last": 2165, "engine": null, "section": "production", "declared_engine": "vllm", "declared_path": "vllm/benchmarks/serve.py", "reviewed_symbol": "default_percentile_metrics", "resolved_engine": "vllm", "resolved_path": "vllm/benchmarks/serve.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/benchmarks/serve.py", "symbol": "default_percentile_metrics", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L2165-L2165" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "serve.py", "first": 661, "last": 666, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/benchmarks/serve.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2465, "page": "10-benchmarking/10-02-benchmark-harnesses.html", "section": "pitfalls", "engine": "vllm", "path": "vllm/benchmarks/serve.py", "first": 661, "last": 666 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L661-L666" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "endpoint_request_func.py", "first": 245, "last": 247, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/benchmarks/lib/endpoint_request_func.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 2443, "page": "10-benchmarking/10-02-benchmark-harnesses.html", "section": "production", "engine": "vllm", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 224, "last": 247 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L245-L247" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/datasets/utils.py", "first": 72, "last": 75, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/utils.py#L72-L75" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/datasets/common.py", "first": 56, "last": 64, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/common.py#L56-L64" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/datasets/utils.py", "first": 70, "last": 79, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/utils.py#L70-L79" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/datasets/common.py", "first": 56, "last": 64, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/common.py#L56-L64" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 559, "last": 572, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L559-L572" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/datasets/random.py", "first": 128, "last": 134, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/random.py#L128-L134" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/one_batch.py", "first": 831, "last": 862, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L831-L862" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/latency.py", "first": 74, "last": 77, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L74-L77" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 2125, "last": 2128, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L2125-L2128" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 31, "last": 61, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L31-L61" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 714, "last": 720, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L714-L720" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 118, "last": 131, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L118-L131" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 661, "last": 666, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L661-L666" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 1693, "last": 1697, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1693-L1697" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/entrypoints/cli/benchmark/serve.py", "first": 11, "last": 22, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/cli/benchmark/serve.py#L11-L22" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/entrypoints/cli/main.py", "first": 57, "last": 57, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/cli/main.py#L57-L57" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/entrypoints/cli/benchmark/main.py", "first": 23, "last": 35, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/cli/benchmark/main.py#L23-L35" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/entrypoints/cli/benchmark/main.py", "first": 38, "last": 46, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/cli/benchmark/main.py#L38-L46" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 1168, "last": 1219, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1168-L1219" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 1615, "last": 1724, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1615-L1724" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/serve.py", "first": 805, "last": 822, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L805-L822" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 71, "last": 82, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L71-L82" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 348, "last": 348, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L348-L348" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/one_batch.py", "first": 856, "last": 862, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L856-L862" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 1, "last": 4, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1-L4" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "python/sglang/benchmark/serving.py", "first": 1378, "last": 1379, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1378-L1379" }, { "page": "10-benchmarking/10-02-benchmark-harnesses.html", "path": "vllm/benchmarks/throughput.py", "first": 1116, "last": 1124, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/throughput.py#L1116-L1124" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 1657, "last": 1665, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1657-L1665" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 1692, "last": 1697, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1692-L1697" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/benchmark/serving.py", "first": 1457, "last": 1466, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1457-L1466" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 725, "last": 734, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L725-L734" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 963, "last": 973, "engine": "vllm", "section": "open-vs-closed", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L963-L973" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 496, "last": 504, "engine": "vllm", "section": "open-vs-closed", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L496-L504" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/benchmark/serving.py", "first": 1081, "last": 1092, "engine": "sglang", "section": "open-vs-closed", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1081-L1092" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 720, "last": 732, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L720-L732" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/model_executor/warmup/kernel_warmup.py", "first": 144, "last": 152, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/warmup/kernel_warmup.py#L144-L152" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 847, "last": 858, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L847-L858" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 966, "last": 981, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L966-L981" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 4345, "last": 4373, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L4345-L4373" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/envs.py", "first": 1419, "last": 1421, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1419-L1421" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/entrypoints/launchers/api_server/routers.py", "first": 34, "last": 37, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/launchers/api_server/routers.py#L34-L37" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/entrypoints/serve/dev/cache/api_router.py", "first": 20, "last": 44, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/dev/cache/api_router.py#L20-L44" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/benchmark/serving.py", "first": 977, "last": 995, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L977-L995" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/sweep/server.py", "first": 15, "last": 142, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/sweep/server.py#L15-L142" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/entrypoints/cli/benchmark/main.py", "first": 23, "last": 35, "engine": "vllm", "section": "confounders", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/cli/benchmark/main.py#L23-L35" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 592, "last": 603, "engine": "vllm", "section": "confounders", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L592-L603" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/benchmark/serving.py", "first": 1122, "last": 1129, "engine": "sglang", "section": "confounders", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1122-L1129" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 1899, "last": 1905, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1899-L1905" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 2214, "last": 2240, "engine": "vllm", "section": "defensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L2214-L2240" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/benchmark/serving.py", "first": 1779, "last": 1790, "engine": "sglang", "section": "defensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1779-L1790" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/utils/jit_monitor.py", "first": 126, "last": 130, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/jit_monitor.py#L126-L130" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "vllm/benchmarks/serve.py", "first": 1678, "last": 1686, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1678-L1686" }, { "page": "10-benchmarking/10-03-honest-benchmarking.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 290, "last": 294, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L290-L294" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "tests/v1/determinism/test_batch_invariance.py", "first": 458, "last": 470, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/test_batch_invariance.py#L458-L470" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/test/test_deterministic.py", "first": 461, "last": 471, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/test/test_deterministic.py#L461-L471" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/kernels/ops/attention/metadata.py", "first": 36, "last": 56, "engine": "sglang", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/metadata.py#L36-L56" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/v1/attention/ops/triton_unified_attention.py", "first": 1035, "last": 1049, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/ops/triton_unified_attention.py#L1035-L1049" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 592, "last": 593, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L592-L593" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/attention/triton_backend.py", "first": 377, "last": 386, "engine": "sglang", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/triton_backend.py#L377-L386" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/attention/triton_backend.py", "first": 286, "last": 296, "engine": "sglang", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/triton_backend.py#L286-L296" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "first": 405, "last": 419, "engine": "sglang", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L405-L419" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/batch_invariant.py", "first": 924, "last": 929, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/batch_invariant.py#L924-L929" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/linear.py", "first": 216, "last": 218, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L216-L218" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/vocab_parallel_embedding.py", "first": 73, "last": 75, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/vocab_parallel_embedding.py#L73-L75" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1307, "last": 1315, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1307-L1315" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/server_args.py", "first": 8351, "last": 8364, "engine": "sglang", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L8351-L8364" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh", "first": 370, "last": 379, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/w8a8/cutlass/c3x/scaled_mm_sm90_fp8_dispatch.cuh#L370-L379" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/kernels/linear/scaled_mm/marlin.py", "first": 44, "last": 45, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/scaled_mm/marlin.py#L44-L45" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/quantization/unquant.py", "first": 120, "last": 125, "engine": "sglang", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/unquant.py#L120-L125" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py", "first": 133, "last": 137, "engine": "vllm", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/router/grouped_topk_router.py#L133-L137" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/moe/topk.py", "first": 1009, "last": 1014, "engine": "sglang", "section": "culprits", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/topk.py#L1009-L1014" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/batch_invariant_ops/__init__.py", "first": 1, "last": 1, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/batch_invariant_ops/__init__.py#L1-L1" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/batch_invariant.py", "first": 951, "last": 973, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/batch_invariant.py#L951-L973" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/batch_invariant_ops/batch_invariant_ops.py", "first": 990, "last": 1005, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/batch_invariant_ops/batch_invariant_ops.py#L990-L1005" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/batch_invariant.py", "first": 341, "last": 445, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/batch_invariant.py#L341-L445" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "csrc/core/batch_invariant.hpp", "first": 9, "last": 16, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/core/batch_invariant.hpp", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/core/batch_invariant.hpp", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/core/batch_invariant.hpp#L9-L16" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/batch_invariant.py", "first": 976, "last": 995, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/batch_invariant.py#L976-L995" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/fused_moe/modular_kernel.py", "first": 677, "last": 684, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/modular_kernel.py#L677-L684" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/v1/attention/selector.py", "first": 243, "last": 247, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L243-L247" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1406, "last": 1410, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1406-L1410" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/batch_invariant.py", "first": 997, "last": 1006, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/batch_invariant.py#L997-L1006" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/envs.py", "first": 627, "last": 628, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L627-L628" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "tests/v1/determinism/utils.py", "first": 26, "last": 36, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/determinism/utils.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/determinism/utils.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/utils.py#L26-L36" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/envs.py", "first": 629, "last": 629, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L629-L629" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/config/vllm.py", "first": 1740, "last": 1748, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L1740-L1748" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/config/parallel.py", "first": 1027, "last": 1029, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L1027-L1029" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/attention/attention.py", "first": 363, "last": 378, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/attention/attention.py#L363-L378" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/v1/attention/selector.py", "first": 169, "last": 169, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/selector.py#L169-L169" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/v1/attention/backends/fa_utils.py", "first": 164, "last": 164, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L164-L164" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/v1/attention/backends/flash_attn.py", "first": 592, "last": 593, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flash_attn.py#L592-L593" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/linear.py", "first": 216, "last": 218, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/linear.py#L216-L218" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vocab_parallel_embedding.py", "first": 73, "last": 75, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/model_executor/layers/vocab_parallel_embedding.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2516, "page": "10-benchmarking/10-04-batch-invariance.html", "section": "culprits", "engine": "vllm", "path": "vllm/model_executor/layers/vocab_parallel_embedding.py", "first": 73, "last": 75 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/vocab_parallel_embedding.py#L73-L75" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/model_executor/model_runner.py", "first": 761, "last": 765, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner.py#L761-L765" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/server_args.py", "first": 8260, "last": 8366, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L8260-L8366" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 1393, "last": 1402, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1393-L1402" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1541, "last": 1557, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1541-L1557" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "vllm/model_executor/layers/fused_moe/fused_moe.py", "first": 1126, "last": 1129, "engine": "vllm", "section": "cost", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/fused_moe/fused_moe.py#L1126-L1129" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/environ.py", "first": 1115, "last": 1115, "engine": "sglang", "section": "cost", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L1115-L1115" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "server_args.py", "first": 8261, "last": 8271, "engine": null, "section": "who-needs-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2546, "page": "10-benchmarking/10-04-batch-invariance.html", "section": "worked-trace", "engine": "sglang", "path": "python/sglang/srt/server_args.py", "first": 8260, "last": 8366 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L8261-L8271" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/sampler.py", "first": 688, "last": 720, "engine": "sglang", "section": "who-needs-it", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L688-L720" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/test/test_deterministic_utils.py", "first": 13, "last": 18, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/test/test_deterministic_utils.py#L13-L18" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 652, "last": 657, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L652-L657" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/sampler.py", "first": 195, "last": 199, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/sampler.py#L195-L199" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/logprob_processor.py", "first": 453, "last": 458, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/logprob_processor.py#L453-L458" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/test/test_deterministic.py", "first": 461, "last": 471, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/test/test_deterministic.py#L461-L471" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "schedule_policy.py", "first": 1393, "last": 1402, "engine": null, "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2547, "page": "10-benchmarking/10-04-batch-invariance.html", "section": "worked-trace", "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 1393, "last": 1402 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1393-L1402" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/kernels/ops/attention/metadata.py", "first": 36, "last": 56, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/attention/metadata.py#L36-L56" }, { "page": "10-benchmarking/10-04-batch-invariance.html", "path": "python/sglang/srt/layers/attention/flashinfer_backend.py", "first": 402, "last": 404, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashinfer_backend.py#L402-L404" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/metrics/perf.py", "first": 4, "last": 7, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/perf.py#L4-L7" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1764, "last": 1766, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1764-L1766" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/metrics/perf.py", "first": 670, "last": 681, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/perf.py#L670-L681" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/config/profiler.py", "first": 42, "last": 52, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/profiler.py#L42-L52" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1146, "last": 1153, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1146-L1153" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/entrypoints/serve/profile/api_router.py", "first": 21, "last": 34, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/profile/api_router.py#L21-L34" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/profiler/wrapper.py", "first": 215, "last": 232, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L215-L232" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/profiler/wrapper.py", "first": 97, "last": 129, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L97-L129" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/profiler.py", "first": 51, "last": 65, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/profiler.py#L51-L65" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/srt/managers/io_struct.py", "first": 2073, "last": 2074, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L2073-L2074" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/srt/managers/scheduler_components/profiler_manager.py", "first": 436, "last": 446, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/profiler_manager.py#L436-L446" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3723, "last": 3724, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3723-L3724" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/benchmarks/latency.py", "first": 125, "last": 129, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L125-L129" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/entrypoints/serve/profile/api_router.py", "first": 41, "last": 45, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/profile/api_router.py#L41-L45" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/srt/utils/profile_utils.py", "first": 476, "last": 487, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/profile_utils.py#L476-L487" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/srt/model_executor/step_span_utils.py", "first": 92, "last": 102, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/step_span_utils.py#L92-L102" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1000, "last": 1024, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1000-L1024" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/entrypoints/serve/profile/api_router.py", "first": 22, "last": 24, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/profile/api_router.py#L22-L24" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/engine/async_llm.py", "first": 945, "last": 949, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L945-L949" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1146, "last": 1199, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1146-L1199" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/profiler/wrapper.py", "first": 85, "last": 95, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L85-L95" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 930, "last": 942, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L930-L942" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/profiler/wrapper.py", "first": 309, "last": 332, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L309-L332" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/profiler/wrapper.py", "first": 472, "last": 489, "engine": "vllm", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L472-L489" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/benchmark/one_batch.py", "first": 121, "last": 126, "engine": "sglang", "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L121-L126" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "test/registered/profiling/test_start_profile.py", "first": 233, "last": 241, "engine": null, "section": "production", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "test/registered/profiling/test_start_profile.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "test/registered/profiling/test_start_profile.py", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/registered/profiling/test_start_profile.py#L233-L241" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/metrics/perf.py", "first": 747, "last": 751, "engine": "vllm", "section": "mla-flop-bug", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/perf.py#L747-L751" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "tests/v1/metrics/test_perf_metrics.py", "first": 1056, "last": 1087, "engine": null, "section": "mla-flop-bug", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/metrics/test_perf_metrics.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/metrics/test_perf_metrics.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/metrics/test_perf_metrics.py#L1056-L1087" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/profiler/wrapper.py", "first": 215, "last": 232, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L215-L232" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 4178, "last": 4183, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4178-L4183" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "vllm/v1/metrics/perf.py", "first": 677, "last": 678, "engine": "vllm", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/perf.py#L677-L678" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "python/sglang/profiler.py", "first": 117, "last": 123, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/profiler.py#L117-L123" }, { "page": "10-benchmarking/10-05-profiling-and-capacity.html", "path": "test/registered/profiling/test_start_profile.py", "first": 144, "last": 252, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "test/registered/profiling/test_start_profile.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "test/registered/profiling/test_start_profile.py", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/registered/profiling/test_start_profile.py#L144-L252" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/gpu/README.md", "first": 1, "last": 4, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/README.md#L1-L4" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/design/vllm_ir.md", "first": 3, "last": 11, "engine": null, "section": "the-repo-map", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/vllm_ir.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/vllm_ir.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/vllm_ir.md#L3-L11" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/usage/v1_guide.md", "first": 9, "last": 15, "engine": "vllm", "section": "rewrite-one", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/usage/v1_guide.md#L9-L15" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/design/model_runner_v2.md", "first": 17, "last": 17, "engine": null, "section": "rewrite-one", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/model_runner_v2.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/model_runner_v2.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/model_runner_v2.md#L17-L17" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/engine/llm_engine.py", "first": 1, "last": 7, "engine": "vllm", "section": "rewrite-one", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/llm_engine.py#L1-L7" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/engine/core.py", "first": 583, "last": 601, "engine": "vllm", "section": "rewrite-one", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L583-L601" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/engine/core.py", "first": 1391, "last": 1402, "engine": "vllm", "section": "rewrite-one", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1391-L1402" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/engine/core.py", "first": 1204, "last": 1204, "engine": "vllm", "section": "rewrite-one", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1204-L1204" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/design/arch_overview.md", "first": 105, "last": 113, "engine": null, "section": "rewrite-one", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/arch_overview.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/arch_overview.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/arch_overview.md#L105-L113" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/design/model_runner_v2.md", "first": 5, "last": 7, "engine": "vllm", "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/model_runner_v2.md#L5-L7" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/design/model_runner_v2.md", "first": 167, "last": 167, "engine": null, "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/model_runner_v2.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/model_runner_v2.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/model_runner_v2.md#L167-L167" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 501, "last": 504, "engine": "vllm", "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L501-L504" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 159, "last": 160, "engine": "vllm", "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L159-L160" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/speculative.py", "first": 219, "last": 225, "engine": "vllm", "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L219-L225" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/gpu/spec_decode/rejection_sampler.py", "first": 84, "last": 97, "engine": "vllm", "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/rejection_sampler.py#L84-L97" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "gpu_input_batch.py", "first": 34, "last": 54, "engine": null, "section": "rewrite-two", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_input_batch.py", "reviewed_symbol": "CachedRequestState", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_input_batch.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_input_batch.py", "symbol": "CachedRequestState", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L34-L54" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/gpu/states.py", "first": 31, "last": 39, "engine": "vllm", "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/states.py#L31-L39" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/design/model_runner_v2.md", "first": 33, "last": 39, "engine": null, "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/model_runner_v2.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/model_runner_v2.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/model_runner_v2.md#L33-L39" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1201, "last": 1236, "engine": "vllm", "section": "rewrite-two", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1201-L1236" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "async_scheduler.py", "first": 46, "last": 49, "engine": null, "section": "rewrite-two", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/async_scheduler.py", "reviewed_symbol": "use_v2_model_runner", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/async_scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/async_scheduler.py", "symbol": "use_v2_model_runner", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/async_scheduler.py#L46-L49" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "v1/attention/backends/flashinfer.py", "first": 940, "last": 940, "engine": null, "section": "rewrite-two", "declared_engine": "vllm", "declared_path": "vllm/v1/attention/backends/flashinfer.py", "reviewed_symbol": "pin_memory", "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/flashinfer.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/attention/backends/flashinfer.py", "symbol": "pin_memory", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L940-L940" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/envs.py", "first": 297, "last": 297, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L297-L297" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/envs.py", "first": 2041, "last": 2044, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L2041-L2044" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/envs.py", "first": 2041, "last": 2044, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L2041-L2044" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 648, "last": 652, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L652" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 654, "last": 680, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L654-L680" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 69, "last": 82, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L69-L82" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 725, "last": 743, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L725-L743" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 685, "last": 689, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L685-L689" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 2438, "last": 2543, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L2438-L2543" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 423, "last": 438, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L423-L438" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/xpu_worker.py", "first": 130, "last": 130, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/xpu_worker.py#L130-L130" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/v1/worker/cpu_worker.py", "first": 174, "last": 174, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/cpu_worker.py#L174-L174" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/entrypoints/openai/api_server.py", "first": 24, "last": 30, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/api_server.py#L24-L30" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 648, "last": 700, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L700" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 2466, "last": 2468, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L2466-L2468" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "vllm/config/vllm.py", "first": 743, "last": 743, "engine": "vllm", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L743-L743" }, { "page": "11-vllm-deep-dive/11-01-repo-map-and-v1.html", "path": "docs/usage/v1_guide.md", "first": 5, "last": 5, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/usage/v1_guide.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/usage/v1_guide.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/usage/v1_guide.md#L5-L5" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/exceptions.py", "first": 12, "last": 21, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/exceptions.py#L12-L21" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/config/parallel.py", "first": 980, "last": 981, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L980-L981" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/uniproc_executor.py", "first": 52, "last": 73, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/uniproc_executor.py#L52-L73" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/input_processor.py", "first": 262, "last": 279, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/input_processor.py#L262-L279" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1091, "last": 1116, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1091-L1116" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/__init__.py", "first": 284, "last": 297, "engine": "vllm", "section": "wire-format", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/__init__.py#L284-L297" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1534, "last": 1548, "engine": "vllm", "section": "wire-format", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1534-L1548" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/__init__.py", "first": 107, "last": 122, "engine": "vllm", "section": "wire-format", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/__init__.py#L107-L122" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/__init__.py", "first": 253, "last": 281, "engine": "vllm", "section": "wire-format", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/__init__.py#L253-L281" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/serial_utils.py", "first": 257, "last": 273, "engine": "vllm", "section": "wire-format", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/serial_utils.py#L257-L273" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/envs.py", "first": 224, "last": 224, "engine": "vllm", "section": "wire-format", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L224-L224" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1829, "last": 1844, "engine": "vllm", "section": "wire-format", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1829-L1844" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/async_llm.py", "first": 135, "last": 156, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L135-L156" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/structured_output/__init__.py", "first": 71, "last": 80, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L71-L80" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/output_processor.py", "first": 48, "last": 55, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L48-L55" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/async_llm.py", "first": 686, "last": 747, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L686-L747" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/async_llm.py", "first": 686, "last": 707, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L686-L707" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/envs.py", "first": 169, "last": 169, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L169-L169" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core_client.py", "first": 88, "last": 110, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L88-L110" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core_client.py", "first": 78, "last": 87, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L78-L87" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core_client.py", "first": 316, "last": 329, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L316-L329" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core_client.py", "first": 1041, "last": 1095, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L1041-L1095" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1390, "last": 1402, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1390-L1402" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 583, "last": 613, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L583-L613" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 310, "last": 310, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L310-L310" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/uniproc_executor.py", "first": 108, "last": 117, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/uniproc_executor.py#L108-L117" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1763, "last": 1772, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1763-L1772" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/config/scheduler.py", "first": 148, "last": 148, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L148-L148" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/config/vllm.py", "first": 1279, "last": 1328, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L1279-L1328" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/uniproc_executor.py", "first": 155, "last": 158, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/uniproc_executor.py#L155-L158" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/abstract.py", "first": 223, "last": 235, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L223-L235" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/abstract.py", "first": 183, "last": 185, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L183-L185" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 315, "last": 441, "engine": "vllm", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L315-L441" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/abstract.py", "first": 61, "last": 81, "engine": "vllm", "section": "executors", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L61-L81" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/uniproc_executor.py", "first": 161, "last": 184, "engine": "vllm", "section": "executors", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/uniproc_executor.py#L161-L184" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/abstract.py", "first": 79, "last": 80, "engine": "vllm", "section": "executors", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L79-L80" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/config/parallel.py", "first": 980, "last": 981, "engine": "vllm", "section": "executors", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L980-L981" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/uniproc_executor.py", "first": 52, "last": 73, "engine": "vllm", "section": "executors", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/uniproc_executor.py#L52-L73" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 1029, "last": 1045, "engine": "vllm", "section": "executors", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L1029-L1045" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 415, "last": 420, "engine": "vllm", "section": "executors", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L415-L420" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 132, "last": 136, "engine": "vllm", "section": "startup", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L132-L136" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 143, "last": 169, "engine": "vllm", "section": "startup", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L143-L169" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/__init__.py", "first": 72, "last": 78, "engine": "vllm", "section": "startup", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/__init__.py#L72-L78" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/async_llm.py", "first": 749, "last": 758, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L749-L758" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/output_processor.py", "first": 508, "last": 529, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L508-L529" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/async_llm.py", "first": 615, "last": 621, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/async_llm.py#L615-L621" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 305, "last": 324, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L305-L324" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1028, "last": 1030, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1028-L1030" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1548, "last": 1549, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1548-L1549" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1618, "last": 1631, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1618-L1631" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core_client.py", "first": 490, "last": 493, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L490-L493" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/output_processor.py", "first": 468, "last": 474, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/output_processor.py#L468-L474" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1477, "last": 1497, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1477-L1497" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/config/vllm.py", "first": 451, "last": 451, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L451-L451" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core_client.py", "first": 428, "last": 440, "engine": "vllm", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core_client.py#L428-L440" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 423, "last": 438, "engine": "vllm", "section": "v2-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L423-L438" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1285, "last": 1289, "engine": "vllm", "section": "reading-guide", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1285-L1289" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/llm_engine.py", "first": 122, "last": 125, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/llm_engine.py#L122-L125" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/outputs.py", "first": 30, "last": 41, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/outputs.py#L30-L41" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/engine/core.py", "first": 1070, "last": 1070, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1070-L1070" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/envs.py", "first": 157, "last": 157, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L157-L157" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/uniproc_executor.py", "first": 180, "last": 183, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/uniproc_executor.py#L180-L183" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/config/parallel.py", "first": 932, "last": 934, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L932-L934" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "abstract.py", "first": 79, "last": 80, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/executor/abstract.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2667, "page": "11-vllm-deep-dive/11-02-engine-layers.html", "section": "executors", "engine": "vllm", "path": "vllm/v1/executor/abstract.py", "first": 79, "last": 80 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L79-L80" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/abstract.py", "first": 116, "last": 117, "engine": "vllm", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L116-L117" }, { "page": "11-vllm-deep-dive/11-02-engine-layers.html", "path": "vllm/v1/executor/abstract.py", "first": 79, "last": 80, "engine": "vllm", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L79-L80" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 484, "last": 495, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L484-L495" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 502, "last": 510, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L502-L510" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 708, "last": 710, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L708-L710" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/config/scheduler.py", "first": 56, "last": 61, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L56-L61" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/config/scheduler.py", "first": 42, "last": 49, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L42-L49" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/engine/arg_utils.py", "first": 2580, "last": 2638, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2580-L2638" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 531, "last": 534, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L531-L534" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 566, "last": 576, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L566-L576" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 630, "last": 636, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L630-L636" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 636, "last": 648, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L636-L648" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 347, "last": 360, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L347-L360" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 685, "last": 700, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L685-L700" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 755, "last": 769, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L755-L769" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1136, "last": 1145, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1136-L1145" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 944, "last": 946, "engine": "vllm", "section": "schedule-walk", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L944-L946" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/request_queue.py", "first": 75, "last": 84, "engine": "vllm", "section": "data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/request_queue.py#L75-L84" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/request.py", "first": 350, "last": 361, "engine": "vllm", "section": "data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L350-L361" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/request.py", "first": 364, "last": 380, "engine": "vllm", "section": "data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L364-L380" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/output.py", "first": 207, "last": 222, "engine": "vllm", "section": "data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/output.py#L207-L222" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "scheduler.py", "first": 1506, "last": 1516, "engine": null, "section": "data-structures", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/scheduler.py", "reviewed_symbol": "async_scheduling", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/scheduler.py", "symbol": "async_scheduling", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1506-L1516" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/engine/__init__.py", "first": 196, "last": 238, "engine": "vllm", "section": "data-structures", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/__init__.py#L196-L238" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1177, "last": 1189, "engine": "vllm", "section": "invariants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1177-L1189" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1803, "last": 1813, "engine": "vllm", "section": "update-from-output", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1803-L1813" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/engine/arg_utils.py", "first": 2619, "last": 2628, "engine": "vllm", "section": "update-from-output", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2619-L2628" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "sched/utils.py", "first": 94, "last": 130, "engine": null, "section": "update-from-output", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/utils.py", "reviewed_symbol": "check_stop", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/utils.py", "symbol": "check_stop", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/utils.py#L94-L130" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1996, "last": 2008, "engine": "vllm", "section": "update-from-output", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1996-L2008" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2060, "last": 2066, "engine": "vllm", "section": "update-from-output", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2060-L2066" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "sched/utils.py", "first": 62, "last": 91, "engine": null, "section": "update-from-output", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/utils.py", "reviewed_symbol": "remove_all", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/utils.py", "symbol": "remove_all", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/utils.py#L62-L91" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/config/scheduler.py", "first": 170, "last": 191, "engine": "vllm", "section": "extension-points", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L170-L191" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/config/scheduler.py", "first": 148, "last": 148, "engine": "vllm", "section": "extension-points", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L148-L148" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/config/vllm.py", "first": 1279, "last": 1328, "engine": "vllm", "section": "extension-points", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L1279-L1328" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/engine/core.py", "first": 148, "last": 148, "engine": "vllm", "section": "extension-points", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L148-L148" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "async_scheduler.py", "first": 19, "last": 19, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/async_scheduler.py", "reviewed_symbol": "_update_after_schedule", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/async_scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/async_scheduler.py", "symbol": "_update_after_schedule", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/async_scheduler.py#L19-L19" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "async_scheduler.py", "first": 39, "last": 41, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/async_scheduler.py", "reviewed_symbol": "request", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/async_scheduler.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/async_scheduler.py", "symbol": "request", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/async_scheduler.py#L39-L41" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/config/vllm.py", "first": 743, "last": 743, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L743-L743" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "request.py", "first": 168, "last": 171, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "vllm/v1/request.py", "reviewed_symbol": "Tokens", "resolved_engine": "vllm", "resolved_path": "vllm/v1/request.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/request.py", "symbol": "Tokens", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L168-L171" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1179, "last": 1179, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1179-L1179" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "utils.py", "first": 84, "last": 89, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/utils.py", "reviewed_symbol": "items_to_remove", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/utils.py", "symbol": "items_to_remove", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/utils.py#L84-L89" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "output.py", "first": 132, "last": 134, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/core/sched/output.py", "reviewed_symbol": "request", "resolved_engine": "vllm", "resolved_path": "vllm/v1/core/sched/output.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/core/sched/output.py", "symbol": "request", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/output.py#L132-L134" }, { "page": "11-vllm-deep-dive/11-03-the-v1-scheduler-in-code.html", "path": "request.py", "first": 386, "last": 387, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/v1/request.py", "reviewed_symbol": "is_finished", "resolved_engine": "vllm", "resolved_path": "vllm/v1/request.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/request.py", "symbol": "is_finished", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/request.py#L386-L387" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/README.md", "first": 1, "last": 4, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/README.md#L1-L4" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2040, "last": 2056, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2040-L2056" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2245, "last": 2262, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2245-L2262" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu_model_runner.py", "first": 2042, "last": 2042, "engine": null, "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 2739, "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "section": "first-principles", "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2040, "last": 2056 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2042-L2042" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/block_table.py", "first": 201, "last": 229, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L201-L229" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/block_table.py", "first": 412, "last": 476, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L412-L476" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/block_table.py", "first": 89, "last": 110, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L89-L110" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 818, "last": 827, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L818-L827" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/block_table.py", "first": 231, "last": 232, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L231-L232" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/vllm.py", "first": 648, "last": 700, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L700" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/envs.py", "first": 2041, "last": 2044, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L2041-L2044" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/vllm.py", "first": 725, "last": 743, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L725-L743" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/vllm.py", "first": 69, "last": 82, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L69-L82" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/vllm.py", "first": 685, "last": 700, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L685-L700" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/vllm.py", "first": 2438, "last": 2543, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L2438-L2543" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 423, "last": 438, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L423-L438" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/core/sched/async_scheduler.py", "first": 46, "last": 46, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/async_scheduler.py#L46-L46" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/attention/backends/flashinfer.py", "first": 940, "last": 940, "engine": "vllm", "section": "which-runner", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/flashinfer.py#L940-L940" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 127, "last": 230, "engine": "vllm", "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L127-L230" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 708, "last": 723, "engine": "vllm", "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L708-L723" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/states.py", "first": 27, "last": 133, "engine": "vllm", "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/states.py#L27-L133" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1057, "last": 1109, "engine": "vllm", "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1057-L1109" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "docs/design/model_runner_v2.md", "first": 36, "last": 42, "engine": null, "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/model_runner_v2.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/model_runner_v2.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/model_runner_v2.md#L36-L42" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 2012, "last": 2024, "engine": "vllm", "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L2012-L2024" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/states.py", "first": 31, "last": 38, "engine": "vllm", "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/states.py#L31-L38" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/buffer_utils.py", "first": 114, "last": 205, "engine": "vllm", "section": "the-input-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/buffer_utils.py#L114-L205" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/attn_utils.py", "first": 659, "last": 680, "engine": "vllm", "section": "attention-and-dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/attn_utils.py#L659-L680" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2355, "last": 2691, "engine": "vllm", "section": "attention-and-dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2355-L2691" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 4000, "last": 4007, "engine": "vllm", "section": "attention-and-dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4000-L4007" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 4054, "last": 4163, "engine": "vllm", "section": "attention-and-dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4054-L4163" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1636, "last": 1645, "engine": "vllm", "section": "attention-and-dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1636-L1645" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "docs/design/model_runner_v2.md", "first": 188, "last": 190, "engine": null, "section": "attention-and-dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/model_runner_v2.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/model_runner_v2.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/model_runner_v2.md#L188-L190" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/cudagraph_utils.py", "first": 382, "last": 410, "engine": "vllm", "section": "attention-and-dispatch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/cudagraph_utils.py#L382-L410" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 501, "last": 503, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L501-L503" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 159, "last": 159, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L159-L159" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "docs/design/model_runner_v2.md", "first": 25, "last": 29, "engine": null, "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/model_runner_v2.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/model_runner_v2.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/model_runner_v2.md#L25-L29" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/input_batch.py", "first": 41, "last": 115, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/input_batch.py#L41-L115" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1361, "last": 1361, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1361-L1361" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 2648, "last": 2668, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2648-L2668" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/speculative.py", "first": 80, "last": 80, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L80-L80" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/spec_decode/rejection_sampler.py", "first": 84, "last": 97, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/rejection_sampler.py#L84-L97" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/sample/rejection_sampler.py", "first": 77, "last": 90, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/rejection_sampler.py#L77-L90" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/speculative.py", "first": 1380, "last": 1395, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L1380-L1395" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/engine/arg_utils.py", "first": 1637, "last": 1640, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L1637-L1640" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/pool/pooling_runner.py", "first": 20, "last": 42, "engine": "vllm", "section": "v1-vs-v2", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/pool/pooling_runner.py#L20-L42" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu/model_runner.py", "first": 1416, "last": 1416, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu/model_runner.py", "reviewed_symbol": "execute_model", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu/model_runner.py", "symbol": "execute_model", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1416-L1416" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu/model_runner.py", "first": 1425, "last": 1432, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1425-L1432" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu/cudagraph_utils.py", "first": 382, "last": 382, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/cudagraph_utils.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2770, "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "section": "attention-and-dispatch", "engine": "vllm", "path": "vllm/v1/worker/gpu/cudagraph_utils.py", "first": 382, "last": 410 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/cudagraph_utils.py#L382-L382" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu/attn_utils.py", "first": 591, "last": 591, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu/attn_utils.py", "reviewed_symbol": "build_attn_metadata", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/attn_utils.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu/attn_utils.py", "symbol": "build_attn_metadata", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/attn_utils.py#L591-L591" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu/input_batch.py", "first": 604, "last": 604, "engine": null, "section": "worked-trace", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu/input_batch.py", "reviewed_symbol": "post_update", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/input_batch.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu/input_batch.py", "symbol": "post_update", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/input_batch.py#L604-L604" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 4321, "last": 4327, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4321-L4327" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 4362, "last": 4371, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L4362-L4371" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "block_table.py", "first": 431, "last": 440, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/block_table.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2743, "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "section": "first-principles", "engine": "vllm", "path": "vllm/v1/worker/block_table.py", "first": 412, "last": 476 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L431-L440" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu_model_runner.py", "first": 2019, "last": 2354, "engine": null, "section": "hands-on", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu_model_runner.py", "reviewed_symbol": "_prepare_inputs", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu_model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu_model_runner.py", "symbol": "_prepare_inputs", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L2019-L2354" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu/model_runner.py", "first": 1110, "last": 1310, "engine": null, "section": "hands-on", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu/model_runner.py", "reviewed_symbol": "prepare_inputs", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/model_runner.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu/model_runner.py", "symbol": "prepare_inputs", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/model_runner.py#L1110-L1310" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu/input_batch.py", "first": 254, "last": 714, "engine": null, "section": "hands-on", "declared_engine": "vllm", "declared_path": "vllm/v1/worker/gpu/input_batch.py", "reviewed_symbol": "_prepare_prefill_inputs_kernel", "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/input_batch.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/v1/worker/gpu/input_batch.py", "symbol": "_prepare_prefill_inputs_kernel", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/input_batch.py#L254-L714" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/vllm.py", "first": 648, "last": 700, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L700" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 734, "last": 830, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L734-L830" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/worker/block_table.py", "first": 412, "last": 476, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/block_table.py#L412-L476" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/config/vllm.py", "first": 743, "last": 743, "engine": "vllm", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L743-L743" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "gpu/spec_decode/rejection_sampler.py", "first": 96, "last": 97, "engine": null, "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/worker/gpu/spec_decode/rejection_sampler.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2778, "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "section": "v1-vs-v2", "engine": "vllm", "path": "vllm/v1/worker/gpu/spec_decode/rejection_sampler.py", "first": 84, "last": 97 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/rejection_sampler.py#L96-L97" }, { "page": "11-vllm-deep-dive/11-04-model-runner-and-backends.html", "path": "vllm/v1/outputs.py", "first": 309, "last": 397, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/outputs.py#L309-L397" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/attention/backends/registry.py", "first": 141, "last": 146, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L141-L146" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/__init__.py", "first": 21, "last": 23, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L21-L23" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/platforms/__init__.py", "first": 229, "last": 229, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/__init__.py#L229-L229" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/__init__.py", "first": 16, "last": 33, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L16-L33" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/__init__.py", "first": 62, "last": 74, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L62-L74" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/envs.py", "first": 1149, "last": 1156, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L1149-L1156" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/__init__.py", "first": 120, "last": 131, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L120-L131" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/engine/arg_utils.py", "first": 811, "last": 811, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L811-L811" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/engine/core.py", "first": 118, "last": 118, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L118-L118" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/worker/worker_base.py", "first": 251, "last": 251, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/worker_base.py#L251-L251" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/models/registry.py", "first": 1531, "last": 1531, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/registry.py#L1531-L1531" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/__init__.py", "first": 77, "last": 90, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L77-L90" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/engine/arg_utils.py", "first": 2882, "last": 2890, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2882-L2890" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/config/model.py", "first": 1246, "last": 1246, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L1246-L1246" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/engine/arg_utils.py", "first": 190, "last": 204, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L190-L204" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/platforms/__init__.py", "first": 229, "last": 229, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/__init__.py#L229-L229" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/endpoint_plugins/interface.py", "first": 5, "last": 16, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/endpoint_plugins/interface.py#L5-L16" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "tests/plugins/vllm_add_dummy_platform/setup.py", "first": 6, "last": 18, "engine": "vllm", "section": "plugin-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/plugins/vllm_add_dummy_platform/setup.py#L6-L18" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/attention/backend.py", "first": 72, "last": 96, "engine": "vllm", "section": "attention-backend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L72-L96" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/attention/backend.py", "first": 744, "last": 762, "engine": "vllm", "section": "attention-backend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L744-L762" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/attention/backend.py", "first": 672, "last": 684, "engine": "vllm", "section": "attention-backend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L672-L684" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/attention/backend.py", "first": 655, "last": 669, "engine": "vllm", "section": "attention-backend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backend.py#L655-L669" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/attention/backends/registry.py", "first": 281, "last": 295, "engine": "vllm", "section": "attention-backend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L281-L295" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/attention/backends/registry.py", "first": 127, "last": 130, "engine": "vllm", "section": "attention-backend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L127-L130" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 84, "last": 103, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L84-L103" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 177, "last": 180, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L177-L180" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/layers/quantization/base_config.py", "first": 179, "last": 192, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/base_config.py#L179-L192" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/kernels/linear/base.py", "first": 191, "last": 209, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/base.py#L191-L209" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/kernels/linear/__init__.py", "first": 649, "last": 663, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L649-L663" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/kernels/linear/__init__.py", "first": 577, "last": 581, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L577-L581" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/kernels/linear/__init__.py", "first": 1127, "last": 1130, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L1127-L1130" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/kernels/linear/__init__.py", "first": 650, "last": 650, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/__init__.py#L650-L650" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/engine/arg_utils.py", "first": 1626, "last": 1626, "engine": "vllm", "section": "quantization", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L1626-L1626" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/entrypoints/openai/completion/protocol.py", "first": 224, "last": 230, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/completion/protocol.py#L224-L230" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/sampling_params.py", "first": 345, "last": 348, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/sampling_params.py#L345-L348" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/interface.py", "first": 71, "last": 108, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/interface.py#L71-L108" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/interface.py", "first": 36, "last": 57, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/interface.py#L36-L57" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/sampler.py", "first": 400, "last": 402, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L400-L402" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/sampler.py", "first": 281, "last": 284, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L281-L284" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 852, "last": 858, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L852-L858" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/worker/gpu_input_batch.py", "first": 961, "last": 961, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_input_batch.py#L961-L961" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/sampler.py", "first": 99, "last": 101, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L99-L101" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/builtin.py", "first": 94, "last": 117, "engine": "vllm", "section": "sampling-param", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/builtin.py#L94-L117" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/core/sched/interface.py", "first": 55, "last": 61, "engine": "vllm", "section": "scheduler-and-connector", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/interface.py#L55-L61" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/core/sched/interface.py", "first": 97, "last": 105, "engine": "vllm", "section": "scheduler-and-connector", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/interface.py#L97-L105" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/engine/core.py", "first": 148, "last": 148, "engine": "vllm", "section": "scheduler-and-connector", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L148-L148" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/base.py", "first": 7, "last": 41, "engine": "vllm", "section": "scheduler-and-connector", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/base.py#L7-L41" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "first": 67, "last": 76, "engine": "vllm", "section": "scheduler-and-connector", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/factory.py#L67-L76" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "first": 30, "last": 40, "engine": "vllm", "section": "scheduler-and-connector", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/factory.py#L30-L40" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/custom_op.py", "first": 196, "last": 207, "engine": "vllm", "section": "custom-ops", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/custom_op.py#L196-L207" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/custom_op.py", "first": 285, "last": 293, "engine": "vllm", "section": "custom-ops", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/custom_op.py#L285-L293" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/custom_op.py", "first": 313, "last": 327, "engine": "vllm", "section": "custom-ops", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/custom_op.py#L313-L327" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/custom_op.py", "first": 338, "last": 346, "engine": "vllm", "section": "custom-ops", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/custom_op.py#L338-L346" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/custom_op.py", "first": 109, "last": 128, "engine": "vllm", "section": "custom-ops", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/custom_op.py#L109-L128" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "tests/plugins/vllm_add_dummy_platform/vllm_add_dummy_platform/dummy_custom_ops.py", "first": 9, "last": 19, "engine": "vllm", "section": "custom-ops", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/plugins/vllm_add_dummy_platform/vllm_add_dummy_platform/dummy_custom_ops.py#L9-L19" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/custom_op.py", "first": 32, "last": 101, "engine": "vllm", "section": "custom-ops", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/custom_op.py#L32-L101" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/config/parallel.py", "first": 259, "last": 261, "engine": "vllm", "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L259-L261" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/__init__.py", "first": 38, "last": 46, "engine": "vllm", "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/__init__.py#L38-L46" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "docs/design/plugin_system.md", "first": 150, "last": 152, "engine": "vllm", "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/plugin_system.md#L150-L152" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/spec_decode/custom_class_proposer.py", "first": 12, "last": 21, "engine": "vllm", "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/custom_class_proposer.py#L12-L21" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "tests/v1/logits_processors/utils.py", "first": 57, "last": 88, "engine": "vllm", "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/logits_processors/utils.py#L57-L88" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/interface.py", "first": 61, "last": 67, "engine": "vllm", "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/interface.py#L61-L67" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/builtin.py", "first": 289, "last": 327, "engine": "vllm", "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/builtin.py#L289-L327" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "docs/features/custom_logitsprocs.md", "first": 347, "last": 348, "engine": "vllm", "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/features/custom_logitsprocs.md#L347-L348" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/__init__.py", "first": 48, "last": 48, "engine": "vllm", "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/__init__.py#L48-L48" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/__init__.py", "first": 79, "last": 84, "engine": "vllm", "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/__init__.py#L79-L84" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/config/model.py", "first": 362, "last": 364, "engine": "vllm", "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L362-L364" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "tests/v1/logits_processors/test_custom_offline.py", "first": 47, "last": 70, "engine": null, "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/logits_processors/test_custom_offline.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/logits_processors/test_custom_offline.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/logits_processors/test_custom_offline.py#L47-L70" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "tests/v1/logits_processors/test_custom_offline.py", "first": 30, "last": 43, "engine": null, "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/logits_processors/test_custom_offline.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/logits_processors/test_custom_offline.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/logits_processors/test_custom_offline.py#L30-L43" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": ".buildkite/test_areas/misc.yaml", "first": 28, "last": 30, "engine": null, "section": "worked-example", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": ".buildkite/test_areas/misc.yaml", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": ".buildkite/test_areas/misc.yaml", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/.buildkite/test_areas/misc.yaml#L28-L30" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "tests/plugins_tests/test_scheduler_plugins.py", "first": 12, "last": 28, "engine": "vllm", "section": "workflow", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/plugins_tests/test_scheduler_plugins.py#L12-L28" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": ".buildkite/test_areas/plugins.yaml", "first": 5, "last": 18, "engine": "vllm", "section": "workflow", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/.buildkite/test_areas/plugins.yaml#L5-L18" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": ".buildkite/test_areas/misc.yaml", "first": 5, "last": 33, "engine": null, "section": "workflow", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": ".buildkite/test_areas/misc.yaml", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": ".buildkite/test_areas/misc.yaml", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/.buildkite/test_areas/misc.yaml#L5-L33" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/__init__.py", "first": 106, "last": 110, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L106-L110" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "examples/features/logits_processor/README.md", "first": 35, "last": 35, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "examples/features/logits_processor/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "examples/features/logits_processor/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/features/logits_processor/README.md#L35-L35" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "examples/features/logits_processor/custom.py", "first": 7, "last": 8, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "examples/features/logits_processor/custom.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "examples/features/logits_processor/custom.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/features/logits_processor/custom.py#L7-L8" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "custom.py", "first": 50, "last": 50, "engine": null, "section": "pitfalls", "declared_engine": "vllm", "declared_path": "examples/features/logits_processor/custom.py", "reviewed_symbol": "DummyLogitsProcessor", "resolved_engine": "vllm", "resolved_path": "examples/features/logits_processor/custom.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "examples/features/logits_processor/custom.py", "symbol": "DummyLogitsProcessor", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/examples/features/logits_processor/custom.py#L50-L50" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/config/scheduler.py", "first": 181, "last": 187, "engine": "vllm", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L181-L187" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/plugins/__init__.py", "first": 18, "last": 30, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L18-L30" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/v1/sample/logits_processor/__init__.py", "first": 48, "last": 48, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/logits_processor/__init__.py#L48-L48" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "factory.py", "first": 33, "last": 34, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2848, "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "section": "scheduler-and-connector", "engine": "vllm", "path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "first": 30, "last": 40 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/factory.py#L33-L34" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "sampler.py", "first": 283, "last": 284, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/sample/sampler.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2838, "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "section": "sampling-param", "engine": "vllm", "path": "vllm/v1/sample/sampler.py", "first": 281, "last": 284 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/sample/sampler.py#L283-L284" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "registry.py", "first": 130, "last": 130, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/v1/attention/backends/registry.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2823, "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "section": "attention-backend", "engine": "vllm", "path": "vllm/v1/attention/backends/registry.py", "first": 127, "last": 130 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L130-L130" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "vllm/model_executor/models/registry.py", "first": 1113, "last": 1119, "engine": "vllm", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/registry.py#L1113-L1119" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "python/sglang/srt/models/registry.py", "first": 24, "last": 35, "engine": "sglang", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L24-L35" }, { "page": "11-vllm-deep-dive/11-05-vllm-extension-points.html", "path": "docs/design/plugin_system.md", "first": 48, "last": 48, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/plugin_system.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/plugin_system.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/plugin_system.md#L48-L48" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 1075, "last": 1079, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1075-L1079" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 892, "last": 919, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L892-L919" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/server_args.py", "first": 9955, "last": 9972, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9955-L9972" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/server_args.py", "first": 10026, "last": 10036, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L10026-L10036" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/scheduler_components/ipc_channels.py", "first": 16, "last": 22, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/ipc_channels.py#L16-L22" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/scheduler_components/ipc_channels.py", "first": 36, "last": 39, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/ipc_channels.py#L36-L39" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/scheduler_components/ipc_channels.py", "first": 52, "last": 65, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/ipc_channels.py#L52-L65" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 547, "last": 556, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L547-L556" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 113, "last": 123, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L113-L123" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 167, "last": 175, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L167-L175" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/data_parallel_controller.py", "first": 150, "last": 154, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/data_parallel_controller.py#L150-L154" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 1179, "last": 1183, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1179-L1183" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/rust_server.py", "first": 1, "last": 8, "engine": "sglang", "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/rust_server.py#L1-L8" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "rust/README.md", "first": 3, "last": 3, "engine": null, "section": "process-architecture", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "rust/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "rust/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/rust/README.md#L3-L3" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/README.md", "first": 5, "last": 11, "engine": "sglang", "section": "repo-map", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/README.md#L5-L11" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/kernels/README.md", "first": 15, "last": 24, "engine": "sglang", "section": "repo-map", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/README.md#L15-L24" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/models/registry.py", "first": 95, "last": 99, "engine": "sglang", "section": "conventions", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L95-L99" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/models/registry.py", "first": 111, "last": 120, "engine": "sglang", "section": "conventions", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L111-L120" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/models/llama.py", "first": 930, "last": 935, "engine": "sglang", "section": "conventions", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/llama.py#L930-L935" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/server_args.py", "first": 461, "last": 472, "engine": "sglang", "section": "conventions", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L461-L472" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/runtime_context.py", "first": 31, "last": 38, "engine": "sglang", "section": "conventions", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/runtime_context.py#L31-L38" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 423, "last": 438, "engine": "vllm", "section": "in-place", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L423-L438" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 384, "last": 392, "engine": "sglang", "section": "in-place", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L384-L392" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/model_executor/runner/__init__.py", "first": 1, "last": 6, "engine": "sglang", "section": "in-place", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/runner/__init__.py#L1-L6" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/server_args.py", "first": 1079, "last": 1104, "engine": "sglang", "section": "parallelism-axes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1079-L1104" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/layers/moe/dwdp/__init__.py", "first": 1, "last": 1, "engine": "sglang", "section": "parallelism-axes", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/moe/dwdp/__init__.py#L1-L1" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 553, "last": 555, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L553-L555" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1748, "last": 1748, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1748-L1748" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/tp_worker.py", "first": 574, "last": 574, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tp_worker.py#L574-L574" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 167, "last": 175, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L167-L175" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 2155, "last": 2155, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L2155-L2155" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/models/registry.py", "first": 104, "last": 109, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L104-L109" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 1095, "last": 1095, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1095-L1095" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 1204, "last": 1210, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1204-L1210" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/scheduler_components/ipc_channels.py", "first": 52, "last": 65, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/ipc_channels.py#L52-L65" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 872, "last": 874, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L872-L874" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "python/sglang/srt/managers/data_parallel_controller.py", "first": 150, "last": 154, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/data_parallel_controller.py#L150-L154" }, { "page": "12-sglang-deep-dive/12-01-repo-map-and-processes.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 429, "last": 429, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L429-L429" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/utils/watchdog.py", "first": 133, "last": 162, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/watchdog.py#L133-L162" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/invariant_checker.py", "first": 462, "last": 485, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/invariant_checker.py#L462-L485" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "vllm/v1/executor/abstract.py", "first": 78, "last": 81, "engine": "vllm", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L78-L81" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/request_receiver.py", "first": 75, "last": 102, "engine": "sglang", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/request_receiver.py#L75-L102" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/request_receiver.py", "first": 209, "last": 216, "engine": "sglang", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/request_receiver.py#L209-L216" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/utils/common.py", "first": 2405, "last": 2420, "engine": "sglang", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L2405-L2420" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/tp_worker.py", "first": 369, "last": 383, "engine": "sglang", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tp_worker.py#L369-L383" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/utils/rank_consensus_checker.py", "first": 406, "last": 432, "engine": "sglang", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/rank_consensus_checker.py#L406-L432" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 495, "last": 499, "engine": "sglang", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L495-L499" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 1753, "last": 1753, "engine": "sglang", "section": "replication", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L1753-L1753" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 215, "last": 245, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L215-L245" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 547, "last": 567, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L547-L567" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 806, "last": 817, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L806-L817" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/ipc_channels.py", "first": 36, "last": 40, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/ipc_channels.py#L36-L40" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/ipc_channels.py", "first": 69, "last": 73, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/ipc_channels.py#L69-L73" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/output_sender.py", "first": 12, "last": 18, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_sender.py#L12-L18" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/tp_worker.py", "first": 584, "last": 596, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tp_worker.py#L584-L596" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/tp_worker.py", "first": 627, "last": 646, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tp_worker.py#L627-L646" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/model_executor/model_runner.py", "first": 1657, "last": 1666, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner.py#L1657-L1666" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/detokenizer_manager.py", "first": 167, "last": 175, "engine": "sglang", "section": "layers", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/detokenizer_manager.py#L167-L175" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/server_args.py", "first": 9955, "last": 9976, "engine": "sglang", "section": "topology", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9955-L9976" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/io_struct.py", "first": 81, "last": 103, "engine": "sglang", "section": "topology", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L81-L103" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/io_struct.py", "first": 1395, "last": 1440, "engine": "sglang", "section": "topology", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L1395-L1440" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/data_parallel_controller.py", "first": 156, "last": 168, "engine": "sglang", "section": "topology", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/data_parallel_controller.py#L156-L168" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 2011, "last": 2037, "engine": "sglang", "section": "schedule-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L2011-L2037" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/model_executor/forward_batch_info.py", "first": 379, "last": 396, "engine": "sglang", "section": "schedule-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/forward_batch_info.py#L379-L396" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 3147, "last": 3175, "engine": "sglang", "section": "schedule-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L3147-L3175" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 3233, "last": 3247, "engine": "sglang", "section": "schedule-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L3233-L3247" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3188, "last": 3197, "engine": "sglang", "section": "schedule-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3188-L3197" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 3299, "last": 3305, "engine": "sglang", "section": "schedule-batch", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L3299-L3305" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1747, "last": 1778, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1747-L1778" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1782, "last": 1800, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1782-L1800" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1828, "last": 1852, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1828-L1852" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/batch_result_processor.py", "first": 805, "last": 814, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/batch_result_processor.py#L805-L814" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/overlap_utils.py", "first": 85, "last": 111, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/overlap_utils.py#L85-L111" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/overlap_utils.py", "first": 246, "last": 249, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/overlap_utils.py#L246-L249" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1857, "last": 1893, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1857-L1893" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/arg_groups/speculative_hook.py", "first": 14, "last": 21, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/speculative_hook.py#L14-L21" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/server_args.py", "first": 4423, "last": 4426, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4423-L4426" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 5014, "last": 5041, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L5014-L5041" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3775, "last": 3775, "engine": "sglang", "section": "overlap", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3775-L3775" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 1060, "last": 1075, "engine": "sglang", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1060-L1075" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 5182, "last": 5189, "engine": "sglang", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L5182-L5189" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 4547, "last": 4586, "engine": "sglang", "section": "lifecycles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L4547-L4586" }, { "page": "12-sglang-deep-dive/12-02-sglang-request-path.html", "path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "first": 165, "last": 169, "engine": "sglang", "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_streamer.py#L165-L169" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "hiradix_cache.py", "first": 77, "last": 77, "engine": null, "section": "the-problem", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/hiradix_cache.py", "reviewed_symbol": "HiRadixCache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/hiradix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/hiradix_cache.py", "symbol": "HiRadixCache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hiradix_cache.py#L77-L77" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 335, "last": 350, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L335-L350" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 235, "last": 235, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L235-L235" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/README.md", "first": 28, "last": 36, "engine": "sglang", "section": "the-file-map", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L28-L36" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/README.md", "first": 38, "last": 45, "engine": "sglang", "section": "the-file-map", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L38-L45" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/base_prefix_cache.py", "first": 315, "last": 335, "engine": "sglang", "section": "the-interface", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/base_prefix_cache.py#L315-L335" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/base_prefix_cache.py", "first": 273, "last": 279, "engine": "sglang", "section": "the-interface", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/base_prefix_cache.py#L273-L279" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "managers/schedule_batch.py", "first": 1362, "last": 1362, "engine": null, "section": "the-interface", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_batch.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3035, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "worked-trace", "engine": "sglang", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1362, "last": 1372 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1362-L1362" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "managers/schedule_policy.py", "first": 156, "last": 156, "engine": null, "section": "the-interface", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/schedule_policy.py", "reviewed_symbol": "match_result", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "symbol": "match_result", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L156-L156" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "managers/schedule_policy.py", "first": 958, "last": 958, "engine": null, "section": "the-interface", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/schedule_policy.py", "reviewed_symbol": "result", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "symbol": "result", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L958-L958" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "managers/schedule_policy.py", "first": 1068, "last": 1068, "engine": null, "section": "the-interface", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/schedule_policy.py", "reviewed_symbol": "tree_cache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "symbol": "tree_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1068-L1068" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "disaggregation/decode.py", "first": 399, "last": 399, "engine": null, "section": "the-interface", "declared_engine": "sglang", "declared_path": "python/sglang/srt/disaggregation/decode.py", "reviewed_symbol": "tree_cache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/disaggregation/decode.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/disaggregation/decode.py", "symbol": "tree_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/decode.py#L399-L399" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "mem_cache/common.py", "first": 131, "last": 131, "engine": null, "section": "the-interface", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/common.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 2985, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-interface", "engine": "sglang", "path": "python/sglang/srt/mem_cache/common.py", "first": 114, "last": 138 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/common.py#L131-L131" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "mem_cache/allocation.py", "first": 261, "last": 261, "engine": null, "section": "the-interface", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/allocation.py", "reviewed_symbol": "tree_cache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/allocation.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/allocation.py", "symbol": "tree_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L261-L261" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "mem_cache/common.py", "first": 111, "last": 111, "engine": null, "section": "the-interface", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/common.py", "reviewed_symbol": "tree_cache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/common.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/common.py", "symbol": "tree_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/common.py#L111-L111" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "mem_cache/common.py", "first": 215, "last": 215, "engine": null, "section": "the-interface", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/common.py", "reviewed_symbol": "tree_cache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/common.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/common.py", "symbol": "tree_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/common.py#L215-L215" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/common.py", "first": 114, "last": 138, "engine": "sglang", "section": "the-interface", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/common.py#L114-L138" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/base_prefix_cache.py", "first": 284, "last": 297, "engine": "sglang", "section": "the-interface", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/base_prefix_cache.py#L284-L297" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/base_prefix_cache.py", "first": 200, "last": 211, "engine": "sglang", "section": "the-interface", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/base_prefix_cache.py#L200-L211" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 199, "last": 199, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L199-L199" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/kv_cache_builder.py", "first": 337, "last": 337, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/kv_cache_builder.py#L337-L337" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 80, "last": 104, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L80-L104" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 104, "last": 142, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L104-L142" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 159, "last": 167, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L159-L167" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 186, "last": 196, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L186-L196" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/component_type.py", "first": 6, "last": 12, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/component_type.py#L6-L12" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "registry.py", "first": 80, "last": 142, "engine": null, "section": "the-factory", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/registry.py", "reviewed_symbol": "default_radix_cache_factory", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "symbol": "default_radix_cache_factory", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L80-L142" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/chunk_cache.py", "first": 35, "last": 77, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/chunk_cache.py#L35-L77" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/environ.py", "first": 588, "last": 588, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L588-L588" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 216, "last": 246, "engine": "sglang", "section": "the-factory", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L216-L246" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 148, "last": 230, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L148-L230" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 104, "last": 136, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L104-L136" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 640, "last": 666, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L640-L666" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 721, "last": 743, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L721-L743" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1074, "last": 1099, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1074-L1099" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 915, "last": 938, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L915-L938" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 518, "last": 536, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L518-L536" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 944, "last": 959, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L944-L959" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1127, "last": 1146, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1127-L1146" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 537, "last": 563, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L537-L563" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "first": 190, "last": 224, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/components/full_component.py#L190-L224" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/evict_policy.py", "first": 41, "last": 65, "engine": "sglang", "section": "the-tree-core", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/evict_policy.py#L41-L65" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 256, "last": 288, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L256-L288" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 1759, "last": 1759, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L1759-L1759" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 2049, "last": 2060, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L2049-L2060" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 3949, "last": 3949, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L3949-L3949" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/allocator/token.py", "first": 42, "last": 64, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/token.py#L42-L64" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/allocator/base.py", "first": 77, "last": 83, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/base.py#L77-L83" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/allocator/paged.py", "first": 191, "last": 219, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/paged.py#L191-L219" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/allocation.py", "first": 332, "last": 346, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L332-L346" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/allocation.py", "first": 348, "last": 361, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L348-L361" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/allocation.py", "first": 69, "last": 85, "engine": "sglang", "section": "the-pools", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L69-L85" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "first": 277, "last": 302, "engine": "sglang", "section": "the-invariants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/components/full_component.py#L277-L302" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1699, "last": 1715, "engine": "sglang", "section": "the-invariants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1699-L1715" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 2271, "last": 2294, "engine": "sglang", "section": "the-invariants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L2271-L2294" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/base_prefix_cache.py", "first": 434, "last": 437, "engine": "sglang", "section": "the-invariants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/base_prefix_cache.py#L434-L437" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/managers/scheduler_components/invariant_checker.py", "first": 453, "last": 459, "engine": "sglang", "section": "the-invariants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/invariant_checker.py#L453-L459" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/managers/scheduler_components/invariant_checker.py", "first": 453, "last": 459, "engine": "sglang", "section": "the-invariants", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/invariant_checker.py#L453-L459" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 495, "last": 495, "engine": null, "section": "rank-consensus", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 3030, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "rank-consensus", "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 495, "last": 514 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L495-L495" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 1753, "last": 1753, "engine": null, "section": "rank-consensus", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "@rank_consensus", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "@rank_consensus", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L1753-L1753" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 1971, "last": 1971, "engine": null, "section": "rank-consensus", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "@rank_consensus", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "@rank_consensus", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L1971-L1971" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 495, "last": 514, "engine": "sglang", "section": "rank-consensus", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L495-L514" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/utils/rank_consensus_checker.py", "first": 87, "last": 88, "engine": "sglang", "section": "rank-consensus", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/rank_consensus_checker.py#L87-L88" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/utils/rank_consensus_checker.py", "first": 413, "last": 431, "engine": "sglang", "section": "rank-consensus", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/rank_consensus_checker.py#L413-L431" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 2122, "last": 2130, "engine": "sglang", "section": "rank-consensus", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L2122-L2130" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 472, "last": 474, "engine": "sglang", "section": "rank-consensus", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L472-L474" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1362, "last": 1372, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1362-L1372" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 495, "last": 495, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3030, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "rank-consensus", "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 495, "last": 514 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L495-L495" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 640, "last": 640, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3001, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-tree-core", "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 640, "last": 666 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L640-L640" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 1059, "last": 1059, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1059-L1059" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 683, "last": 683, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "inc_lock_ref", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "inc_lock_ref", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L683-L683" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 557, "last": 557, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "reviewed_symbol": "inc_lock_ref", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "symbol": "inc_lock_ref", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L557-L557" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "full_component.py", "first": 259, "last": 259, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "reviewed_symbol": "acquire_component_lock", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "symbol": "acquire_component_lock", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/components/full_component.py#L259-L259" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/allocation.py", "first": 281, "last": 281, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L281-L281" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "memory_pool.py", "first": 291, "last": 291, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/memory_pool.py", "reviewed_symbol": "def alloc", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/memory_pool.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/memory_pool.py", "symbol": "def alloc", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L291-L291" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "allocator/token.py", "first": 55, "last": 55, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/allocator/token.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3015, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-pools", "engine": "sglang", "path": "python/sglang/srt/mem_cache/allocator/token.py", "first": 42, "last": 64 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/token.py#L55-L55" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "allocation.py", "first": 54, "last": 54, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/allocation.py", "reviewed_symbol": "write_cache_indices", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/allocation.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/allocation.py", "symbol": "write_cache_indices", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L54-L54" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "allocation.py", "first": 512, "last": 512, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/allocation.py", "reviewed_symbol": "alloc_for_decode", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/allocation.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/allocation.py", "symbol": "alloc_for_decode", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocation.py#L512-L512" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "mem_cache/common.py", "first": 198, "last": 219, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/common.py", "reviewed_symbol": "release_kv_cache", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/common.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/common.py", "symbol": "release_kv_cache", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/common.py#L198-L219" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 743, "last": 743, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "cache_finished_req", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "cache_finished_req", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L743-L743" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 706, "last": 706, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "_dec_req_lock", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "_dec_req_lock", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L706-L706" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "common.py", "first": 114, "last": 114, "engine": null, "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/common.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2985, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-interface", "engine": "sglang", "path": "python/sglang/srt/mem_cache/common.py", "first": 114, "last": 138 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/common.py#L114-L114" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 620, "last": 620, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "_evict_components", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "_evict_components", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L620-L620" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 1245, "last": 1245, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "reviewed_symbol": "evict_device_leaf", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "symbol": "evict_device_leaf", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1245-L1245" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_radix_cache.py", "first": 565, "last": 565, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "reviewed_symbol": "_free_values", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "symbol": "_free_values", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L565-L565" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/unified_radix_cache.py", "first": 896, "last": 919, "engine": "sglang", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_radix_cache.py#L896-L919" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 991, "last": 994, "engine": null, "section": "worked-trace", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "reviewed_symbol": "FreeDeviceKV", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "symbol": "FreeDeviceKV", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L991-L994" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "registry.py", "first": 187, "last": 196, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2993, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-factory", "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "first": 186, "last": 196 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L187-L196" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "schedule_policy.py", "first": 235, "last": 235, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2971, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-problem", "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 235, "last": 235 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L235-L235" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "environ.py", "first": 588, "last": 588, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/environ.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2997, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-factory", "engine": "sglang", "path": "python/sglang/srt/environ.py", "first": 588, "last": 588 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L588-L588" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "registry.py", "first": 106, "last": 106, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2991, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-factory", "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "first": 104, "last": 142 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L106-L106" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "registry.py", "first": 118, "last": 141, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2991, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-factory", "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "first": 104, "last": 142 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L118-L141" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "server_args.py", "first": 3058, "last": 3066, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "enable_flexkv", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "enable_flexkv", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3058-L3066" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "mem_cache/README.md", "first": 67, "last": 72, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/README.md", "reviewed_symbol": "Allocator", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/README.md", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/README.md", "symbol": "Allocator", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L67-L72" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 2400, "last": 2426, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "reviewed_symbol": "evictable_size", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "symbol": "evictable_size", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L2400-L2426" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "registry.py", "first": 248, "last": 256, "engine": null, "section": "hands-on", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/registry.py", "reviewed_symbol": "Tree cache initialized", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "symbol": "Tree cache initialized", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L248-L256" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1367, "last": 1367, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1367-L1367" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "allocator/paged.py", "first": 126, "last": 126, "engine": null, "section": "hands-on", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/allocator/paged.py", "reviewed_symbol": "debug_mode", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/allocator/paged.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/allocator/paged.py", "symbol": "debug_mode", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/paged.py#L126-L126" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 1699, "last": 1715, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3022, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-invariants", "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 1699, "last": 1715 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L1699-L1715" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 2170, "last": 2200, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "reviewed_symbol": "expected_dev_leaves", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "symbol": "expected_dev_leaves", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L2170-L2200" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "memory_pool.py", "first": 279, "last": 282, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/memory_pool.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3011, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-pools", "engine": "sglang", "path": "python/sglang/srt/mem_cache/memory_pool.py", "first": 256, "last": 288 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L279-L282" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "registry.py", "first": 85, "last": 89, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2990, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-factory", "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "first": 80, "last": 104 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L85-L89" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "unified_tree_core.py", "first": 724, "last": 725, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3002, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-tree-core", "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py", "first": 721, "last": 743 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/unified_tree_core.py#L724-L725" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "registry.py", "first": 80, "last": 142, "engine": null, "section": "takeaways", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/registry.py", "reviewed_symbol": "default_radix_cache_factory", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "symbol": "default_radix_cache_factory", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L80-L142" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "python/sglang/srt/mem_cache/README.md", "first": 43, "last": 43, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L43-L43" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "README.md", "first": 5, "last": 6, "engine": null, "section": "further-reading", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/README.md", "reviewed_symbol": "layout", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/README.md", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/README.md", "symbol": "layout", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L5-L6" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "README.md", "first": 44, "last": 45, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/README.md", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 2973, "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "section": "the-file-map", "engine": "sglang", "path": "python/sglang/srt/mem_cache/README.md", "first": 38, "last": 45 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L44-L45" }, { "page": "12-sglang-deep-dive/12-03-radix-cache-in-code.html", "path": "memory_pool.py", "first": 295, "last": 304, "engine": null, "section": "further-reading", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/memory_pool.py", "reviewed_symbol": "20476", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/memory_pool.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/memory_pool.py", "symbol": "20476", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/memory_pool.py#L295-L304" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 199, "last": 210, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L199-L210" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/README.md", "first": 5, "last": 11, "engine": "sglang", "section": "frontend-dsl", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/README.md#L5-L11" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/lang/ir.py", "first": 148, "last": 148, "engine": "sglang", "section": "frontend-dsl", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/ir.py#L148-L148" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/test/test_programs.py", "first": 262, "last": 276, "engine": "sglang", "section": "frontend-dsl", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/test/test_programs.py#L262-L276" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/lang/interpreter.py", "first": 370, "last": 400, "engine": "sglang", "section": "frontend-dsl", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/interpreter.py#L370-L400" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/lang/backend/runtime_endpoint.py", "first": 166, "last": 167, "engine": "sglang", "section": "frontend-dsl", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/backend/runtime_endpoint.py#L166-L167" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/lang/backend/runtime_endpoint.py", "first": 105, "last": 114, "engine": "sglang", "section": "frontend-dsl", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/backend/runtime_endpoint.py#L105-L114" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/lang/interpreter.py", "first": 1052, "last": 1066, "engine": "sglang", "section": "frontend-dsl", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/interpreter.py#L1052-L1066" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 58, "last": 107, "engine": "sglang", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L58-L107" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 241, "last": 257, "engine": "sglang", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L241-L257" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "vllm/v1/structured_output/backend_types.py", "first": 99, "last": 118, "engine": "vllm", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_types.py#L99-L118" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 347, "last": 366, "engine": "sglang", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L347-L366" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "vllm/v1/structured_output/__init__.py", "first": 130, "last": 145, "engine": "vllm", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L130-L145" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "test/registered/unit/constrained/test_base_grammar_backend.py", "first": 218, "last": 223, "engine": "sglang", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/registered/unit/constrained/test_base_grammar_backend.py#L218-L223" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 284, "last": 296, "engine": "sglang", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L284-L296" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/constrained/grammar_manager.py", "first": 184, "last": 197, "engine": "sglang", "section": "grammar-backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/grammar_manager.py#L184-L197" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "base_grammar_backend.py", "first": 109, "last": 110, "engine": null, "section": "grammar-backends", "declared_engine": "sglang", "declared_path": "python/sglang/srt/constrained/base_grammar_backend.py", "reviewed_symbol": "copy", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/constrained/base_grammar_backend.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "symbol": "copy", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L109-L110" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/Cargo.toml", "first": 1, "last": 32, "engine": "sglang", "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/Cargo.toml#L1-L32" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/bindings/python/pyproject.toml", "first": 44, "last": 53, "engine": "sglang", "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/pyproject.toml#L44-L53" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/README.md", "first": 47, "last": 52, "engine": null, "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/README.md#L47-L52" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/router.py", "first": 7, "last": 34, "engine": null, "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/router.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/bindings/python/src/sglang_router/router.py", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/router.py#L7-L34" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/src/service_discovery.rs", "first": 32, "last": 50, "engine": "sglang", "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/service_discovery.rs#L32-L50" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "service_discovery.rs", "first": 57, "last": 57, "engine": null, "section": "the-router", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/service_discovery.rs", "reviewed_symbol": "check_interval", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/service_discovery.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/service_discovery.rs", "symbol": "check_interval", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/service_discovery.rs#L57-L57" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "service_discovery.rs", "first": 63, "last": 63, "engine": null, "section": "the-router", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/src/service_discovery.rs", "reviewed_symbol": "bootstrap_port_annotation", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/service_discovery.rs", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/src/service_discovery.rs", "symbol": "bootstrap_port_annotation", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/service_discovery.rs#L63-L63" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "first": 54, "last": 61, "engine": "sglang", "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L54-L61" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "mini_lb.py", "first": 103, "last": 112, "engine": null, "section": "the-router", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "reviewed_symbol": "select_pair", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "symbol": "select_pair", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L103-L112" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "first": 128, "last": 141, "engine": "sglang", "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L128-L141" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 1178, "last": 1182, "engine": "sglang", "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1178-L1182" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "rust/README.md", "first": 3, "last": 3, "engine": null, "section": "the-router", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "rust/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "rust/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/rust/README.md#L3-L3" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/models/registry.py", "first": 109, "last": 125, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L109-L125" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/models/registry.py", "first": 129, "last": 134, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L129-L134" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/layers/attention/attention_registry.py", "first": 31, "last": 39, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L31-L39" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "attention_registry.py", "first": 41, "last": 66, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/layers/attention/attention_registry.py", "reviewed_symbol": "create_flashinfer_backend", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/attention_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/layers/attention/attention_registry.py", "symbol": "create_flashinfer_backend", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L41-L66" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py", "first": 254, "last": 257, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner_components/attention_backend_setup.py#L254-L257" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "server_args.py", "first": 181, "last": 181, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "ATTENTION_BACKEND_CHOICES", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "ATTENTION_BACKEND_CHOICES", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L181-L181" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "server_args.py", "first": 405, "last": 406, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "add_attention_backend_choices", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "add_attention_backend_choices", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L405-L406" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "server_args.py", "first": 401, "last": 419, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "add_quantization_method_choices", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "add_quantization_method_choices", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L401-L419" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "server_args.py", "first": 211, "last": 214, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "Out-of-tree platforms", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "Out-of-tree platforms", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L211-L214" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/speculative/spec_registry.py", "first": 222, "last": 243, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_registry.py#L222-L243" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "spec_registry.py", "first": 168, "last": 180, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/spec_registry.py", "reviewed_symbol": "_reserved_names", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/spec_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/spec_registry.py", "symbol": "_reserved_names", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_registry.py#L168-L180" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "spec_registry.py", "first": 184, "last": 201, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/speculative/spec_registry.py", "reviewed_symbol": "_assert_custom_spec_algo_conforms", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/speculative/spec_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/speculative/spec_registry.py", "symbol": "_assert_custom_spec_algo_conforms", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_registry.py#L184-L201" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/layers/quantization/__init__.py", "first": 143, "last": 143, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/__init__.py#L143-L143" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/layers/quantization/__init__.py", "first": 163, "last": 170, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/quantization/__init__.py#L163-L170" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "arg_groups/overrides.py", "first": 15, "last": 20, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/arg_groups/overrides.py", "reviewed_symbol": "identity", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/arg_groups/overrides.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/arg_groups/overrides.py", "symbol": "identity", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L15-L20" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 86, "last": 100, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L86-L100" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "overrides.py", "first": 354, "last": 360, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/arg_groups/overrides.py", "reviewed_symbol": "Collect", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/arg_groups/overrides.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/arg_groups/overrides.py", "symbol": "Collect", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L354-L360" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/plugins/hook_registry.py", "first": 60, "last": 66, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/plugins/hook_registry.py#L60-L66" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "hook_registry.py", "first": 15, "last": 21, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/plugins/hook_registry.py", "reviewed_symbol": "result", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/plugins/hook_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/plugins/hook_registry.py", "symbol": "result", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/plugins/hook_registry.py#L15-L21" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/plugins/__init__.py", "first": 27, "last": 29, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/plugins/__init__.py#L27-L29" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "hook_registry.py", "first": 120, "last": 136, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/plugins/hook_registry.py", "reviewed_symbol": "hook_type", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/plugins/hook_registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/plugins/hook_registry.py", "symbol": "hook_type", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/plugins/hook_registry.py#L120-L136" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/entrypoints/engine.py", "first": 238, "last": 240, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L238-L240" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 5124, "last": 5125, "engine": "sglang", "section": "extension-seams", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L5124-L5125" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "engine.py", "first": 1086, "last": 1088, "engine": null, "section": "extension-seams", "declared_engine": "sglang", "declared_path": "python/sglang/srt/entrypoints/engine.py", "reviewed_symbol": "Defensive", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/entrypoints/engine.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/entrypoints/engine.py", "symbol": "Defensive", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/engine.py#L1086-L1088" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/src/policies/factory.rs", "first": 17, "last": 64, "engine": null, "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/factory.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/policies/factory.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/factory.rs#L17-L64" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "interpreter.py", "first": 461, "last": 502, "engine": null, "section": "not-extensible", "declared_engine": "sglang", "declared_path": "python/sglang/lang/interpreter.py", "reviewed_symbol": "_execute", "resolved_engine": "sglang", "resolved_path": "python/sglang/lang/interpreter.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/lang/interpreter.py", "symbol": "_execute", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/interpreter.py#L461-L502" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/cli/serve_backends.py", "first": 155, "last": 163, "engine": "sglang", "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/cli/serve_backends.py#L155-L163" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "serve_backends.py", "first": 24, "last": 24, "engine": null, "section": "not-extensible", "declared_engine": "sglang", "declared_path": "python/sglang/cli/serve_backends.py", "reviewed_symbol": "SERVE_BACKEND_API_VERSION", "resolved_engine": "sglang", "resolved_path": "python/sglang/cli/serve_backends.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/cli/serve_backends.py", "symbol": "SERVE_BACKEND_API_VERSION", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/cli/serve_backends.py#L24-L24" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "developer_guide/serve_backend_plugins.mdx", "first": 149, "last": 149, "engine": null, "section": "not-extensible", "declared_engine": "sglang", "declared_path": "docs/docs/developer_guide/serve_backend_plugins.mdx", "reviewed_symbol": "Declare the API version", "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/serve_backend_plugins.mdx", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "docs/docs/developer_guide/serve_backend_plugins.mdx", "symbol": "Declare the API version", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/serve_backend_plugins.mdx#L149-L149" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/src/server.rs", "first": 615, "last": 617, "engine": "sglang", "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/server.rs#L615-L617" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "sgl-model-gateway/Cargo.toml", "first": 131, "last": 131, "engine": null, "section": "not-extensible", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/Cargo.toml", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/Cargo.toml", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/Cargo.toml#L131-L131" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "test/README.md", "first": 62, "last": 73, "engine": null, "section": "workflow", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "test/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "test/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/README.md#L62-L73" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "developer_guide/contribution_guide.mdx", "first": 44, "last": 50, "engine": null, "section": "workflow", "declared_engine": "sglang", "declared_path": "docs/docs/developer_guide/contribution_guide.mdx", "reviewed_symbol": "registered", "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "docs/docs/developer_guide/contribution_guide.mdx", "symbol": "registered", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L44-L50" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "test/README.md", "first": 10, "last": 10, "engine": null, "section": "workflow", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "test/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "test/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/README.md#L10-L10" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "developer_guide/contribution_guide.mdx", "first": 120, "last": 128, "engine": null, "section": "workflow", "declared_engine": "sglang", "declared_path": "docs/docs/developer_guide/contribution_guide.mdx", "reviewed_symbol": "request", "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "docs/docs/developer_guide/contribution_guide.mdx", "symbol": "request", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L120-L128" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "contribution_guide.mdx", "first": 27, "last": 35, "engine": null, "section": "workflow", "declared_engine": "sglang", "declared_path": "docs/docs/developer_guide/contribution_guide.mdx", "reviewed_symbol": "install", "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "docs/docs/developer_guide/contribution_guide.mdx", "symbol": "install", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L27-L35" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "contribution_guide.mdx", "first": 163, "last": 166, "engine": null, "section": "workflow", "declared_engine": "sglang", "declared_path": "docs/docs/developer_guide/contribution_guide.mdx", "reviewed_symbol": "Minimize", "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "docs/docs/developer_guide/contribution_guide.mdx", "symbol": "Minimize", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L163-L166" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "mini_lb.py", "first": 64, "last": 65, "engine": null, "section": "pitfalls", "declared_engine": "sglang", "declared_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "reviewed_symbol": "PD disaggregation mode", "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py", "symbol": "PD disaggregation mode", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/bindings/python/src/sglang_router/mini_lb.py#L64-L65" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "interpreter.py", "first": 392, "last": 399, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/lang/interpreter.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3081, "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "section": "frontend-dsl", "engine": "sglang", "path": "python/sglang/lang/interpreter.py", "first": 370, "last": 400 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/interpreter.py#L392-L399" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "interpreter.py", "first": 729, "last": 736, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/lang/interpreter.py", "reviewed_symbol": "_execute_concatenate_and_append_text", "resolved_engine": "sglang", "resolved_path": "python/sglang/lang/interpreter.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/lang/interpreter.py", "symbol": "_execute_concatenate_and_append_text", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/lang/interpreter.py#L729-L736" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "base_grammar_backend.py", "first": 360, "last": 360, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/constrained/base_grammar_backend.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3088, "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "section": "grammar-backends", "engine": "sglang", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 347, "last": 366 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L360-L360" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "mem_cache/registry.py", "first": 65, "last": 68, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/mem_cache/registry.py", "reviewed_symbol": "_RADIX_CACHE_REGISTRY", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/registry.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/mem_cache/registry.py", "symbol": "_RADIX_CACHE_REGISTRY", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L65-L68" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "attention_registry.py", "first": 31, "last": 31, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/layers/attention/attention_registry.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3108, "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "section": "extension-seams", "engine": "sglang", "path": "python/sglang/srt/layers/attention/attention_registry.py", "first": 31, "last": 39 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L31-L31" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "server_args.py", "first": 181, "last": 181, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "ATTENTION_BACKEND_CHOICES", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "ATTENTION_BACKEND_CHOICES", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L181-L181" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "server_args.py", "first": 405, "last": 406, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/server_args.py", "reviewed_symbol": "add_attention_backend_choices", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/server_args.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/server_args.py", "symbol": "add_attention_backend_choices", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L405-L406" }, { "page": "12-sglang-deep-dive/12-04-frontend-grammar-router.html", "path": "python/sglang/README.md", "first": 8, "last": 8, "engine": "sglang", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/README.md#L8-L8" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/benchmarks/datasets/utils.py", "first": 66, "last": 73, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/utils.py#L66-L73" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/benchmark/datasets/common.py", "first": 56, "last": 64, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/common.py#L56-L64" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 340, "last": 351, "engine": "vllm", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L340-L351" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/managers/scheduler_components/request_receiver.py", "first": 75, "last": 79, "engine": "sglang", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/request_receiver.py#L75-L79" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 740, "last": 768, "engine": "sglang", "section": "mental-model", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L740-L768" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/config/cache.py", "first": 59, "last": 59, "engine": "vllm", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L59-L59" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 236, "last": 242, "engine": "vllm", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L236-L242" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/core/single_type_kv_cache_manager.py", "first": 516, "last": 524, "engine": "vllm", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/single_type_kv_cache_manager.py#L516-L524" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2394, "last": 2397, "engine": "sglang", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2394-L2397" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/mem_cache/allocator/token.py", "first": 44, "last": 66, "engine": "sglang", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/allocator/token.py#L44-L66" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "first": 190, "last": 198, "engine": "sglang", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/components/full_component.py#L190-L198" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 1347, "last": 1388, "engine": "vllm", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L1347-L1388" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1941, "last": 1945, "engine": "sglang", "section": "where-complexity-lives", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1941-L1945" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/config/cache.py", "first": 59, "last": 59, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L59-L59" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2378, "last": 2397, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2378-L2397" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/core/block_pool.py", "first": 143, "last": 190, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L143-L190" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 236, "last": 242, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L236-L242" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "first": 190, "last": 198, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/components/full_component.py#L190-L198" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/executor/multiproc_executor.py", "first": 340, "last": 351, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/multiproc_executor.py#L340-L351" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 740, "last": 768, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L740-L768" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/executor/abstract.py", "first": 79, "last": 79, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/executor/abstract.py#L79-L79" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/config/vllm.py", "first": 435, "last": 435, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L435-L435" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/config/vllm.py", "first": 1448, "last": 1452, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L1448-L1452" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/model_executor/cuda_graph_config.py", "first": 110, "last": 119, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/cuda_graph_config.py#L110-L119" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/attention/backends/fa_utils.py", "first": 70, "last": 199, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/fa_utils.py#L70-L199" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/layers/attention/flashattention_backend.py", "first": 299, "last": 300, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/flashattention_backend.py#L299-L300" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 1645, "last": 1646, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L1645-L1646" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/config/speculative.py", "first": 150, "last": 151, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L150-L151" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/server_args.py", "first": 6443, "last": 6456, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L6443-L6456" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/distributed/kv_events.py", "first": 50, "last": 119, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_events.py#L50-L119" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 1, "last": 32, "engine": null, "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/cache_aware.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/policies/cache_aware.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L1-L32" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/models/registry.py", "first": 24, "last": 35, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L24-L35" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/layers/attention/attention_registry.py", "first": 31, "last": 38, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L31-L38" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/config/vllm.py", "first": 648, "last": 700, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L648-L700" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/config/vllm.py", "first": 725, "last": 744, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L725-L744" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/mem_cache/README.md", "first": 1, "last": 40, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/README.md#L1-L40" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/worker/sentinel/gpu_worker_sentinel.py", "first": 29, "last": 45, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/sentinel/gpu_worker_sentinel.py#L29-L45" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/utils/watchdog.py", "first": 160, "last": 163, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/watchdog.py#L160-L163" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "rust/README.md", "first": 1, "last": 3, "engine": null, "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "rust/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "rust/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/rust/README.md#L1-L3" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/envs.py", "first": 165, "last": 165, "engine": "vllm", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L165-L165" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/managers/rust_server.py", "first": 1, "last": 8, "engine": "sglang", "section": "the-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/rust_server.py#L1-L8" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 238, "last": 240, "engine": "vllm", "section": "no-benchmark-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L238-L240" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/benchmarks/serve.py", "first": 595, "last": 598, "engine": "vllm", "section": "no-benchmark-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L595-L598" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/benchmark/serving.py", "first": 753, "last": 761, "engine": "sglang", "section": "no-benchmark-table", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L753-L761" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/distributed/kv_events.py", "first": 50, "last": 56, "engine": "vllm", "section": "convergence", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_events.py#L50-L56" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/disaggregation/kv_events.py", "first": 112, "last": 118, "engine": "sglang", "section": "convergence", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/kv_events.py#L112-L118" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "docs/design/paged_attention.md", "first": 3, "last": 5, "engine": null, "section": "convergence", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/design/paged_attention.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/design/paged_attention.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/paged_attention.md#L3-L5" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "docs/design/paged_attention.md", "first": 3, "last": 5, "engine": "vllm", "section": "convergence", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/design/paged_attention.md#L3-L5" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/constrained/reasoner_grammar_backend.py", "first": 226, "last": 229, "engine": "sglang", "section": "convergence", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/reasoner_grammar_backend.py#L226-L229" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 79, "last": 143, "engine": "sglang", "section": "convergence", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L79-L143" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/mem_cache/radix_cache.py", "first": 335, "last": 350, "engine": "sglang", "section": "convergence", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/radix_cache.py#L335-L350" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/core/block_pool.py", "first": 719, "last": 744, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L719-L744" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/server_args.py", "first": 5906, "last": 5906, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5906-L5906" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 236, "last": 242, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L236-L242" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "python/sglang/srt/server_args.py", "first": 5906, "last": 5906, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5906-L5906" }, { "page": "13-comparison-and-frontier/13-01-head-to-head.html", "path": "full_component.py", "first": 190, "last": 198, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3162, "page": "13-comparison-and-frontier/13-01-head-to-head.html", "section": "where-complexity-lives", "engine": "sglang", "path": "python/sglang/srt/mem_cache/unified_cache/components/full_component.py", "first": 190, "last": 198 }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/unified_cache/components/full_component.py#L190-L198" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 4808, "last": 4830, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4808-L4830" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/mem_cache/registry.py", "first": 91, "last": 102, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/registry.py#L91-L102" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/mem_cache/chunk_cache.py", "first": 35, "last": 41, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/chunk_cache.py#L35-L41" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/mem_cache/chunk_cache.py", "first": 67, "last": 77, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/chunk_cache.py#L67-L77" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 843, "last": 858, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L843-L858" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 200, "last": 213, "engine": "sglang", "section": "workload-shape", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L200-L213" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/v1/core/sched/request_queue.py", "first": 13, "last": 17, "engine": "vllm", "section": "workload-shape", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/request_queue.py#L13-L17" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/config/scheduler.py", "first": 70, "last": 72, "engine": "vllm", "section": "workload-shape", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L70-L72" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 816, "last": 820, "engine": "sglang", "section": "workload-shape", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L816-L820" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/model_executor/models/registry.py", "first": 72, "last": 77, "engine": "vllm", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/registry.py#L72-L77" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/models/registry.py", "first": 95, "last": 110, "engine": "sglang", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L95-L110" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/config/model.py", "first": 2086, "last": 2096, "engine": "vllm", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/model.py#L2086-L2096" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 3985, "last": 3993, "engine": "sglang", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3985-L3993" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 3927, "last": 3929, "engine": "sglang", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L3927-L3929" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/platforms/__init__.py", "first": 210, "last": 216, "engine": "vllm", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/__init__.py#L210-L216" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 181, "last": 189, "engine": "sglang", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L181-L189" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/engine/arg_utils.py", "first": 2609, "last": 2638, "engine": "vllm", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2609-L2638" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 4866, "last": 4883, "engine": "sglang", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4866-L4883" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 143, "last": 152, "engine": "sglang", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L143-L152" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 12, "last": 36, "engine": "vllm", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L12-L36" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/model_executor/models/siglip.py", "first": 421, "last": 421, "engine": "vllm", "section": "the-gates", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/models/siglip.py#L421-L421" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 8746, "last": 8749, "engine": "sglang", "section": "team-capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L8746-L8749" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/plugins/__init__.py", "first": 16, "last": 30, "engine": "vllm", "section": "team-capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/plugins/__init__.py#L16-L30" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 397, "last": 454, "engine": "sglang", "section": "team-capacity", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L397-L454" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 573, "last": 577, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L573-L577" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 4380, "last": 4380, "engine": "vllm", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L4380-L4380" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/benchmark/datasets/generated_shared_prefix.py", "first": 23, "last": 41, "engine": "sglang", "section": "hands-on", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/datasets/generated_shared_prefix.py#L23-L41" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 4866, "last": 4883, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L4866-L4883" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "vllm/engine/arg_utils.py", "first": 2609, "last": 2638, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2609-L2638" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/models/registry.py", "first": 95, "last": 110, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/models/registry.py#L95-L110" }, { "page": "13-comparison-and-frontier/13-02-choosing-an-engine.html", "path": "python/sglang/srt/server_args.py", "first": 9333, "last": 9336, "engine": "sglang", "section": "takeaways", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9333-L9336" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py", "first": 688, "last": 705, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py#L688-L705" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/metadata.py", "first": 170, "last": 188, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/metadata.py#L170-L188" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/tp_mapping.py", "first": 79, "last": 84, "engine": "vllm", "section": "first-principles", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/tp_mapping.py#L79-L84" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "first": 152, "last": 241, "engine": "vllm", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/factory.py#L152-L241" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/disaggregation/utils.py", "first": 589, "last": 594, "engine": "sglang", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/utils.py#L589-L594" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py", "first": 627, "last": 633, "engine": "vllm", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py#L627-L633" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py", "first": 497, "last": 502, "engine": "vllm", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py#L497-L502" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py", "first": 657, "last": 658, "engine": "vllm", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py#L657-L658" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 156, "last": 157, "engine": "vllm", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L156-L157" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 2905, "last": 2909, "engine": "vllm", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L2905-L2909" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/disaggregation/decode.py", "first": 877, "last": 893, "engine": "sglang", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/decode.py#L877-L893" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/disaggregation/prefill.py", "first": 956, "last": 961, "engine": "sglang", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/prefill.py#L956-L961" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/disaggregation/decode.py", "first": 1621, "last": 1628, "engine": "sglang", "section": "disaggregation-at-scale", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/disaggregation/decode.py#L1621-L1628" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/mem_cache/hicache_storage.py", "first": 301, "last": 323, "engine": "sglang", "section": "kv-as-distributed-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hicache_storage.py#L301-L323" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/mem_cache/hicache_storage.py", "first": 345, "last": 354, "engine": "sglang", "section": "kv-as-distributed-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hicache_storage.py#L345-L354" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/mem_cache/hicache_storage.py", "first": 380, "last": 388, "engine": "sglang", "section": "kv-as-distributed-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/hicache_storage.py#L380-L388" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 144, "last": 156, "engine": "vllm", "section": "kv-as-distributed-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L144-L156" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/config/cache.py", "first": 109, "last": 120, "engine": "vllm", "section": "kv-as-distributed-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L109-L120" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/mem_cache/storage/file/lru_file_evictor.py", "first": 6, "last": 21, "engine": "sglang", "section": "kv-as-distributed-system", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/mem_cache/storage/file/lru_file_evictor.py#L6-L21" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/v1/attention/backends/registry.py", "first": 44, "last": 131, "engine": "vllm", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L44-L131" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/v1/attention/backends/registry.py", "first": 186, "last": 193, "engine": "vllm", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L186-L193" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/layers/attention/attention_registry.py", "first": 41, "last": 493, "engine": "sglang", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/layers/attention/attention_registry.py#L41-L493" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/platforms/cpu.py", "first": 89, "last": 90, "engine": "vllm", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cpu.py#L89-L90" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/platforms/cpu.py", "first": 215, "last": 220, "engine": "vllm", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cpu.py#L215-L220" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/platforms/cpu.py", "first": 166, "last": 175, "engine": "vllm", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/cpu.py#L166-L175" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/platforms/interface.py", "first": 419, "last": 430, "engine": "vllm", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/interface.py#L419-L430" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/utils/common.py", "first": 968, "last": 971, "engine": "sglang", "section": "heterogeneous-hardware", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L968-L971" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/server_args.py", "first": 5376, "last": 5384, "engine": "sglang", "section": "ultra-long-context", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L5376-L5384" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/configs/model_config.py", "first": 207, "last": 209, "engine": "sglang", "section": "ultra-long-context", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/configs/model_config.py#L207-L209" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 757, "last": 765, "engine": "vllm", "section": "ultra-long-context", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L757-L765" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/platforms/interface.py", "first": 1052, "last": 1062, "engine": "vllm", "section": "on-device", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/platforms/interface.py#L1052-L1062" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/utils/mem_utils.py", "first": 148, "last": 156, "engine": "vllm", "section": "on-device", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/mem_utils.py#L148-L156" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_wNa16.py", "first": 39, "last": 39, "engine": "vllm", "section": "on-device", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_wNa16.py#L39-L39" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/benchmarks/serve.py", "first": 642, "last": 645, "engine": "vllm", "section": "cross-cutting", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L642-L645" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/benchmarks/serve.py", "first": 607, "last": 613, "engine": "vllm", "section": "cross-cutting", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L607-L613" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "tests/v1/kv_connector/nixl_integration/test_accuracy.py", "first": 10, "last": 15, "engine": "vllm", "section": "cross-cutting", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/kv_connector/nixl_integration/test_accuracy.py#L10-L15" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "python/sglang/srt/server_args.py", "first": 6511, "last": 6519, "engine": "sglang", "section": "cross-cutting", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L6511-L6519" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/pull_scheduler.py", "first": 237, "last": 281, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/pull_scheduler.py#L237-L281" }, { "page": "13-comparison-and-frontier/13-03-open-problems.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py", "first": 490, "last": 493, "engine": "vllm", "section": "worked-trace", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/nixl/base_worker.py#L490-L493" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/docs/developer_guide/contribution_guide.mdx", "first": 115, "last": 121, "engine": "sglang", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L115-L121" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/contributing/README.md", "first": 304, "last": 310, "engine": "vllm", "section": "the-problem", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/contributing/README.md#L304-L310" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/core/sched/interface.py", "first": 38, "last": 54, "engine": "vllm", "section": "rebase-debt", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/interface.py#L38-L54" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1146, "last": 1154, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1146-L1154" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 931, "last": 940, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L931-L940" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/config/profiler.py", "first": 105, "last": 109, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/profiler.py#L105-L109" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 944, "last": 961, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L944-L961" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/config/profiler.py", "first": 118, "last": 123, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/profiler.py#L118-L123" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 424, "last": 438, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L424-L438" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "python/sglang/srt/managers/io_struct.py", "first": 2063, "last": 2077, "engine": "sglang", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L2063-L2077" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "python/sglang/srt/managers/scheduler_components/invariant_checker.py", "first": 64, "last": 81, "engine": "sglang", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/invariant_checker.py#L64-L81" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "tests/v1/core/test_prefix_caching.py", "first": 1369, "last": 1381, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/core/test_prefix_caching.py#L1369-L1381" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 106, "last": 106, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L106-L106" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "first": 108, "last": 115, "engine": null, "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "csrc/libtorch_stable/attention/merge_attn_states.cu", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/attention/merge_attn_states.cu#L108-L115" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/metrics/stats.py", "first": 131, "last": 143, "engine": "vllm", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L131-L143" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "python/sglang/srt/managers/scheduler_components/metrics_reporter.py", "first": 672, "last": 685, "engine": "sglang", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/metrics_reporter.py#L672-L685" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 905, "last": 911, "engine": "sglang", "section": "a-weekend", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L905-L911" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/attention/backends/registry.py", "first": 258, "last": 268, "engine": "vllm", "section": "a-week", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/attention/backends/registry.py#L258-L268" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "tests/test_attention_backend_registry.py", "first": 106, "last": 121, "engine": "vllm", "section": "a-week", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/test_attention_backend_registry.py#L106-L121" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/config/scheduler.py", "first": 180, "last": 191, "engine": "vllm", "section": "a-week", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L180-L191" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 200, "last": 215, "engine": "sglang", "section": "a-week", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L200-L215" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/docs/developer_guide/quantization_contribution_guide.mdx", "first": 46, "last": 59, "engine": "sglang", "section": "a-week", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/quantization_contribution_guide.mdx#L46-L59" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/docs/developer_guide/contribution_guide.mdx", "first": 84, "last": 102, "engine": null, "section": "a-week", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/docs/developer_guide/contribution_guide.mdx", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L84-L102" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/spec_decode/custom_class_proposer.py", "first": 12, "last": 21, "engine": "vllm", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/custom_class_proposer.py#L12-L21" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 646, "last": 650, "engine": "vllm", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L646-L650" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/config/vllm.py", "first": 649, "last": 699, "engine": "vllm", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L649-L699" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/worker/gpu/spec_decode/__init__.py", "first": 51, "last": 58, "engine": "vllm", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu/spec_decode/__init__.py#L51-L58" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "python/sglang/srt/speculative/spec_registry.py", "first": 222, "last": 243, "engine": "sglang", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/speculative/spec_registry.py#L222-L243" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/distributed/kv_transfer/kv_connector/factory.py", "first": 30, "last": 41, "engine": "vllm", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/factory.py#L30-L41" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "sgl-model-gateway/src/policies/mod.rs", "first": 41, "last": 55, "engine": "sglang", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/mod.rs#L41-L55" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "sgl-model-gateway/src/policies/factory.rs", "first": 15, "last": 22, "engine": "sglang", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/factory.rs#L15-L22" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 5, "last": 16, "engine": "sglang", "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L5-L16" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "sgl-model-gateway/src/policies/cache_aware.rs", "first": 21, "last": 23, "engine": null, "section": "a-month", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "sgl-model-gateway/src/policies/cache_aware.rs", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "sgl-model-gateway/src/policies/cache_aware.rs", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/sgl-model-gateway/src/policies/cache_aware.rs#L21-L23" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/contributing/README.md", "first": 277, "last": 281, "engine": null, "section": "a-quarter", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/contributing/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/contributing/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/contributing/README.md#L277-L281" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/docs/developer_guide/contribution_guide.mdx", "first": 237, "last": 239, "engine": null, "section": "a-quarter", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/docs/developer_guide/contribution_guide.mdx", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L237-L239" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": ".buildkite/test_areas/spec_decode.yaml", "first": 5, "last": 23, "engine": "vllm", "section": "contributing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/.buildkite/test_areas/spec_decode.yaml#L5-L23" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "test/README.md", "first": 61, "last": 73, "engine": "sglang", "section": "contributing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/README.md#L61-L73" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "test/README.md", "first": 10, "last": 10, "engine": null, "section": "contributing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "test/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "test/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/README.md#L10-L10" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/contributing/deprecation_policy.md", "first": 21, "last": 27, "engine": "vllm", "section": "contributing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/contributing/deprecation_policy.md#L21-L27" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/contributing/README.md", "first": 198, "last": 221, "engine": null, "section": "contributing", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/contributing/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/contributing/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/contributing/README.md#L198-L221" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/docs/developer_guide/contribution_guide.mdx", "first": 163, "last": 166, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/docs/developer_guide/contribution_guide.mdx", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L163-L166" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/docs/developer_guide/quantization_contribution_guide.mdx", "first": 79, "last": 83, "engine": null, "section": "pitfalls", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/quantization_contribution_guide.mdx", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/docs/developer_guide/quantization_contribution_guide.mdx", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/quantization_contribution_guide.mdx#L79-L83" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "schedule_policy.py", "first": 558, "last": 558, "engine": null, "section": "exercises", "declared_engine": "sglang", "declared_path": "python/sglang/srt/managers/schedule_policy.py", "reviewed_symbol": "page alignment", "resolved_engine": "sglang", "resolved_path": "python/sglang/srt/managers/schedule_policy.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "sglang", "path": "python/sglang/srt/managers/schedule_policy.py", "symbol": "page alignment", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L558-L558" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/v1/spec_decode/llm_base_proposer.py", "first": 1850, "last": 1853, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/llm_base_proposer.py#L1850-L1853" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "vllm/distributed/kv_transfer/kv_connector/v1/multi_connector.py", "first": 391, "last": 395, "engine": "vllm", "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/kv_transfer/kv_connector/v1/multi_connector.py#L391-L395" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/contributing/README.md", "first": 15, "last": 22, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "docs/contributing/README.md", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/contributing/README.md", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/contributing/README.md#L15-L22" }, { "page": "13-comparison-and-frontier/13-04-what-to-build-next.html", "path": "docs/docs/developer_guide/contribution_guide.mdx", "first": 233, "last": 239, "engine": null, "section": "further-reading", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "sglang", "resolved_path": "docs/docs/developer_guide/contribution_guide.mdx", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "docs/docs/developer_guide/contribution_guide.mdx", "checked_engines": [ "sglang", "vllm" ], "engine": "sglang", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/docs/docs/developer_guide/contribution_guide.mdx#L233-L239" }, { "page": "index.html", "path": "vllm/v1/core/sched/scheduler.py", "first": 412, "last": 458, "engine": "vllm", "section": "contract", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/sched/scheduler.py#L412-L458" }, { "page": "labs/01-measure-your-gpu-roofline/README.html", "path": "benchmarks/kernels/weight_shapes.py", "first": 52, "last": 57, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/kernels/weight_shapes.py#L52-L57" }, { "page": "labs/01-measure-your-gpu-roofline/README.html", "path": "benchmarks/kernels/benchmark_fp8_gemm.py", "first": 89, "last": 97, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/kernels/benchmark_fp8_gemm.py#L89-L97" }, { "page": "labs/01-measure-your-gpu-roofline/README.html", "path": "benchmarks/kernels/benchmark_cp_gather.py", "first": 193, "last": 201, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/kernels/benchmark_cp_gather.py#L193-L201" }, { "page": "labs/01-measure-your-gpu-roofline/README.html", "path": "vllm/config/observability.py", "first": 77, "last": 78, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/observability.py#L77-L78" }, { "page": "labs/01-measure-your-gpu-roofline/README.html", "path": "vllm/v1/metrics/perf.py", "first": 1521, "last": 1534, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/perf.py#L1521-L1534" }, { "page": "labs/01-measure-your-gpu-roofline/README.html", "path": "python/sglang/srt/managers/scheduler_components/metrics_reporter.py", "first": 892, "last": 903, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/metrics_reporter.py#L892-L903" }, { "page": "labs/02-kv-cache-sizing/README.html", "path": "vllm/config/cache.py", "first": 80, "last": 80, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L80-L80" }, { "page": "labs/02-kv-cache-sizing/README.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1924, "last": 1931, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1924-L1931" }, { "page": "labs/02-kv-cache-sizing/README.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1094, "last": 1100, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1094-L1100" }, { "page": "labs/02-kv-cache-sizing/README.html", "path": "vllm/v1/kv_cache_interface.py", "first": 250, "last": 259, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/kv_cache_interface.py#L250-L259" }, { "page": "labs/03-continuous-batching-effects/README.html", "path": "vllm/benchmarks/serve.py", "first": 607, "last": 613, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L607-L613" }, { "page": "labs/03-continuous-batching-effects/README.html", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 400, "last": 412, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L400-L412" }, { "page": "labs/03-continuous-batching-effects/README.html", "path": "vllm/benchmarks/serve.py", "first": 1588, "last": 1599, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1588-L1599" }, { "page": "labs/03-continuous-batching-effects/README.html", "path": "python/sglang/benchmark/serving.py", "first": 2378, "last": 2390, "engine": "sglang", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L2378-L2390" }, { "page": "labs/03-continuous-batching-effects/README.html", "path": "vllm/v1/metrics/loggers.py", "first": 261, "last": 267, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L261-L267" }, { "page": "labs/03-continuous-batching-effects/README.html", "path": "python/sglang/srt/managers/scheduler_components/metrics_reporter.py", "first": 810, "last": 811, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/metrics_reporter.py#L810-L811" }, { "page": "labs/03-continuous-batching-effects/README.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3619, "last": 3631, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3619-L3631" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 221, "last": 231, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L221-L231" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/core/kv_cache_manager.py", "first": 256, "last": 262, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_manager.py#L256-L262" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 745, "last": 752, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L745-L752" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/config/cache.py", "first": 59, "last": 59, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L59-L59" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2378, "last": 2380, "engine": "sglang", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2378-L2380" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/metrics/loggers.py", "first": 584, "last": 602, "engine": "vllm", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L584-L602" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "tests/v1/kv_connector/nixl_integration/test_mamba_prefix_cache.py", "first": 234, "last": 247, "engine": "vllm", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/kv_connector/nixl_integration/test_mamba_prefix_cache.py#L234-L247" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/metrics/loggers.py", "first": 287, "last": 297, "engine": "vllm", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L287-L297" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/metrics/stats.py", "first": 42, "last": 42, "engine": "vllm", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/stats.py#L42-L42" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 892, "last": 902, "engine": "sglang", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L892-L902" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 292, "last": 297, "engine": "sglang", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L292-L297" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 284, "last": 286, "engine": "sglang", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L284-L286" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/utils/common.py", "first": 2542, "last": 2548, "engine": "sglang", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L2542-L2548" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/managers/scheduler_components/metrics_reporter.py", "first": 604, "last": 612, "engine": "sglang", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/metrics_reporter.py#L604-L612" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1420, "last": 1425, "engine": "sglang", "section": "metrics", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1420-L1425" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/config/cache.py", "first": 107, "last": 108, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L107-L108" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/entrypoints/launchers/api_server/routers.py", "first": 33, "last": 37, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/launchers/api_server/routers.py#L33-L37" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/entrypoints/serve/dev/cache/api_router.py", "first": 20, "last": 24, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/dev/cache/api_router.py#L20-L24" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 966, "last": 976, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L966-L976" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/entrypoints/openai/completion/protocol.py", "first": 50, "last": 56, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/completion/protocol.py#L50-L56" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/managers/io_struct.py", "first": 170, "last": 177, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/io_struct.py#L170-L177" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 618, "last": 622, "engine": "vllm", "section": "break", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L618-L622" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/entrypoints/openai/chat_completion/protocol.py", "first": 467, "last": 478, "engine": "vllm", "section": "break", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/chat_completion/protocol.py#L467-L478" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/v1/core/block_pool.py", "first": 658, "last": 668, "engine": "vllm", "section": "break", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/block_pool.py#L658-L668" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/config/cache.py", "first": 101, "last": 103, "engine": "vllm", "section": "break", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L101-L103" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/srt/server_args.py", "first": 797, "last": 808, "engine": "sglang", "section": "break", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L797-L808" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "vllm/benchmarks/datasets/datasets.py", "first": 1858, "last": 1878, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/datasets/datasets.py#L1858-L1878" }, { "page": "labs/04-prefix-cache-hit-rate/README.html", "path": "python/sglang/benchmark/serving.py", "first": 2610, "last": 2628, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L2610-L2628" }, { "page": "labs/05-chunked-prefill/README.html", "path": "python/sglang/srt/server_args.py", "first": 1480, "last": 1484, "engine": "sglang", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1480-L1484" }, { "page": "labs/05-chunked-prefill/README.html", "path": "vllm/entrypoints/openai/completion/protocol.py", "first": 79, "last": 83, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/completion/protocol.py#L79-L83" }, { "page": "labs/05-chunked-prefill/README.html", "path": "python/sglang/srt/entrypoints/openai/protocol.py", "first": 366, "last": 371, "engine": "sglang", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/protocol.py#L366-L371" }, { "page": "labs/05-chunked-prefill/README.html", "path": "vllm/config/scheduler.py", "first": 250, "last": 261, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L250-L261" }, { "page": "labs/05-chunked-prefill/README.html", "path": "vllm/config/scheduler.py", "first": 241, "last": 246, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L241-L246" }, { "page": "labs/05-chunked-prefill/README.html", "path": "python/sglang/srt/server_args.py", "first": 811, "last": 815, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L811-L815" }, { "page": "labs/05-chunked-prefill/README.html", "path": "python/sglang/srt/server_args.py", "first": 996, "last": 1000, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L996-L1000" }, { "page": "labs/05-chunked-prefill/README.html", "path": "python/sglang/srt/server_args.py", "first": 9286, "last": 9291, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L9286-L9291" }, { "page": "labs/05-chunked-prefill/README.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 3461, "last": 3467, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L3461-L3467" }, { "page": "labs/05-chunked-prefill/README.html", "path": "vllm/config/scheduler.py", "first": 70, "last": 72, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/scheduler.py#L70-L72" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "csrc/libtorch_stable/quantization/w8a8/cutlass/scaled_mm_entry.cu", "first": 145, "last": 159, "engine": "vllm", "section": null, "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/libtorch_stable/quantization/w8a8/cutlass/scaled_mm_entry.cu#L145-L159" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/config/cache.py", "first": 80, "last": 80, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L80-L80" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/model_executor/layers/quantization/__init__.py", "first": 12, "last": 46, "engine": "vllm", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/__init__.py#L12-L46" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/config/quantization.py", "first": 116, "last": 130, "engine": "vllm", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/quantization.py#L116-L130" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/model_executor/layers/quantization/fp8.py", "first": 95, "last": 106, "engine": "vllm", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/fp8.py#L95-L106" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "python/sglang/srt/server_args.py", "first": 143, "last": 164, "engine": "sglang", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L143-L164" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/engine/arg_utils.py", "first": 1221, "last": 1221, "engine": "vllm", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L1221-L1221" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/config/cache.py", "first": 88, "last": 90, "engine": "vllm", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/cache.py#L88-L90" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "python/sglang/srt/server_args.py", "first": 690, "last": 701, "engine": "sglang", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L690-L701" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "test/registered/quant/test_awq.py", "first": 56, "last": 64, "engine": "sglang", "section": "flags", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/test/registered/quant/test_awq.py#L56-L64" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "python/sglang/srt/model_executor/model_runner_components/load_model_utils.py", "first": 156, "last": 164, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner_components/load_model_utils.py#L156-L164" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "python/sglang/srt/model_executor/model_runner.py", "first": 1158, "last": 1166, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/model_executor/model_runner.py#L1158-L1166" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/v1/worker/gpu_model_runner.py", "first": 5510, "last": 5514, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_model_runner.py#L5510-L5514" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/v1/core/kv_cache_utils.py", "first": 1925, "last": 1931, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/core/kv_cache_utils.py#L1925-L1931" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 1094, "last": 1100, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L1094-L1100" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/benchmarks/latency.py", "first": 35, "last": 53, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L35-L53" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "python/sglang/benchmark/one_batch_server.py", "first": 1, "last": 11, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch_server.py#L1-L11" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "tests/evals/gsm8k/README.md", "first": 16, "last": 22, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/README.md#L16-L22" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "tests/evals/gsm8k/gsm8k_eval.py", "first": 342, "last": 361, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/gsm8k_eval.py#L342-L361" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "python/sglang/test/run_eval.py", "first": 1, "last": 4, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/test/run_eval.py#L1-L4" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "tests/evals/gsm8k/configs/Qwen3.5-35B-A3B-DEP2.yaml", "first": 1, "last": 9, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/configs/Qwen3.5-35B-A3B-DEP2.yaml#L1-L9" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "tests/evals/gsm8k/configs/Qwen3.5-35B-A3B-FP8-DEP2.yaml", "first": 1, "last": 10, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/configs/Qwen3.5-35B-A3B-FP8-DEP2.yaml#L1-L10" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "tests/evals/gsm8k/test_gsm8k_correctness.py", "first": 189, "last": 192, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/evals/gsm8k/test_gsm8k_correctness.py#L189-L192" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/model_executor/layers/quantization/fp8.py", "first": 286, "last": 292, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/quantization/fp8.py#L286-L292" }, { "page": "labs/06-quantization-tradeoff/README.html", "path": "vllm/model_executor/kernels/linear/scaled_mm/cutlass.py", "first": 287, "last": 295, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/kernels/linear/scaled_mm/cutlass.py#L287-L295" }, { "page": "labs/07-tp-scaling/README.html", "path": "vllm/benchmarks/latency.py", "first": 46, "last": 53, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L46-L53" }, { "page": "labs/07-tp-scaling/README.html", "path": "vllm/benchmarks/latency.py", "first": 170, "last": 177, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/latency.py#L170-L177" }, { "page": "labs/07-tp-scaling/README.html", "path": "python/sglang/benchmark/one_batch.py", "first": 857, "last": 864, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/one_batch.py#L857-L864" }, { "page": "labs/07-tp-scaling/README.html", "path": "tools/profiler/nsys_profile_tools/vllm_engine_model.json", "first": 3, "last": 19, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tools/profiler/nsys_profile_tools/vllm_engine_model.json#L3-L19" }, { "page": "labs/07-tp-scaling/README.html", "path": "examples/profiler/nsys_profile_tools/sglang_engine_model.json", "first": 3, "last": 20, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/examples/profiler/nsys_profile_tools/sglang_engine_model.json#L3-L20" }, { "page": "labs/07-tp-scaling/README.html", "path": "csrc/custom_all_reduce.cuh", "first": 7, "last": 13, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/csrc/custom_all_reduce.cuh#L7-L13" }, { "page": "labs/07-tp-scaling/README.html", "path": "vllm/distributed/device_communicators/cuda_communicator.py", "first": 287, "last": 289, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/cuda_communicator.py#L287-L289" }, { "page": "labs/07-tp-scaling/README.html", "path": "vllm/distributed/device_communicators/custom_all_reduce.py", "first": 348, "last": 361, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/custom_all_reduce.py#L348-L361" }, { "page": "labs/07-tp-scaling/README.html", "path": "vllm/distributed/device_communicators/all_reduce_utils.py", "first": 31, "last": 37, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/distributed/device_communicators/all_reduce_utils.py#L31-L37" }, { "page": "labs/07-tp-scaling/README.html", "path": "python/sglang/srt/distributed/device_communicators/custom_all_reduce.py", "first": 41, "last": 42, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/distributed/device_communicators/custom_all_reduce.py#L41-L42" }, { "page": "labs/07-tp-scaling/README.html", "path": "vllm/config/parallel.py", "first": 205, "last": 205, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/parallel.py#L205-L205" }, { "page": "labs/07-tp-scaling/README.html", "path": "python/sglang/srt/server_args.py", "first": 2021, "last": 2028, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2021-L2028" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/config/speculative.py", "first": 180, "last": 186, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L180-L186" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "python/sglang/srt/server_args.py", "first": 2278, "last": 2287, "engine": "sglang", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L2278-L2287" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/v1/spec_decode/metrics.py", "first": 177, "last": 196, "engine": "vllm", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/metrics.py#L177-L196" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/v1/spec_decode/metrics.py", "first": 227, "last": 232, "engine": "vllm", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/metrics.py#L227-L232" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/benchmarks/serve.py", "first": 1107, "last": 1119, "engine": "vllm", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1107-L1119" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/benchmarks/serve.py", "first": 1381, "last": 1408, "engine": "vllm", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1381-L1408" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/config/observability.py", "first": 48, "last": 53, "engine": "vllm", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/observability.py#L48-L53" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/entrypoints/openai/engine/protocol.py", "first": 132, "last": 151, "engine": "vllm", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/engine/protocol.py#L132-L151" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "python/sglang/srt/entrypoints/openai/protocol.py", "first": 418, "last": 427, "engine": "sglang", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/protocol.py#L418-L427" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "python/sglang/srt/managers/scheduler_components/metrics_reporter.py", "first": 818, "last": 830, "engine": "sglang", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/metrics_reporter.py#L818-L830" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "python/sglang/srt/managers/scheduler.py", "first": 4404, "last": 4411, "engine": "sglang", "section": "metric", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler.py#L4404-L4411" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "docs/benchmarking/cli.md", "first": 371, "last": 371, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/docs/benchmarking/cli.md#L371-L371" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "python/sglang/benchmark/serving.py", "first": 2266, "last": 2272, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L2266-L2272" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/config/speculative.py", "first": 219, "last": 241, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/speculative.py#L219-L241" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "python/sglang/srt/entrypoints/openai/protocol.py", "first": 418, "last": 427, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/protocol.py#L418-L427" }, { "page": "labs/08-spec-decode-acceptance/README.html", "path": "vllm/v1/spec_decode/utils.py", "first": 598, "last": 601, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/spec_decode/utils.py#L598-L601" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/kernels/ops/grammar/bitmask_ops.py", "first": 65, "last": 77, "engine": "sglang", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/kernels/ops/grammar/bitmask_ops.py#L65-L77" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/config/structured_outputs.py", "first": 12, "last": 24, "engine": "vllm", "section": "backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/structured_outputs.py#L12-L24" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/engine/arg_utils.py", "first": 1672, "last": 1674, "engine": "vllm", "section": "backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L1672-L1674" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/utils/argparse_utils.py", "first": 117, "last": 125, "engine": "vllm", "section": "backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/utils/argparse_utils.py#L117-L125" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/server_args.py", "first": 245, "last": 245, "engine": "sglang", "section": "backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L245-L245" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 415, "last": 424, "engine": "sglang", "section": "backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L415-L424" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/v1/structured_output/backend_guidance.py", "first": 25, "last": 32, "engine": "vllm", "section": "backends", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_guidance.py#L25-L32" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/v1/engine/core.py", "first": 593, "last": 603, "engine": "vllm", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L593-L603" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/v1/structured_output/__init__.py", "first": 61, "last": 69, "engine": "vllm", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L61-L69" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/v1/structured_output/__init__.py", "first": 250, "last": 256, "engine": "vllm", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/__init__.py#L250-L256" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/constrained/base_grammar_backend.py", "first": 88, "last": 94, "engine": "sglang", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/base_grammar_backend.py#L88-L94" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/constrained/llguidance_backend.py", "first": 151, "last": 159, "engine": "sglang", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/llguidance_backend.py#L151-L159" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/sampling/sampling_batch_info.py", "first": 253, "last": 264, "engine": "sglang", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/sampling/sampling_batch_info.py#L253-L264" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/constrained/grammar_manager.py", "first": 204, "last": 225, "engine": "sglang", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/constrained/grammar_manager.py#L204-L225" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/environ.py", "first": 374, "last": 376, "engine": "sglang", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/environ.py#L374-L376" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/observability/metrics_collector.py", "first": 280, "last": 285, "engine": "sglang", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/observability/metrics_collector.py#L280-L285" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/v1/structured_output/backend_xgrammar.py", "first": 66, "last": 71, "engine": "vllm", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/structured_output/backend_xgrammar.py#L66-L71" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "benchmarks/benchmark_serving_structured_output.py", "first": 161, "last": 169, "engine": "vllm", "section": "where", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_serving_structured_output.py#L161-L169" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "benchmarks/benchmark_serving_structured_output.py", "first": 1072, "last": 1077, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_serving_structured_output.py#L1072-L1077" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "benchmarks/benchmark_serving_structured_output.py", "first": 912, "last": 916, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_serving_structured_output.py#L912-L916" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "benchmarks/benchmark_serving_structured_output.py", "first": 484, "last": 489, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_serving_structured_output.py#L484-L489" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "vllm/entrypoints/openai/engine/protocol.py", "first": 218, "last": 227, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/openai/engine/protocol.py#L218-L227" }, { "page": "labs/09-structured-decoding-overhead/README.html", "path": "python/sglang/srt/entrypoints/openai/protocol.py", "first": 1044, "last": 1055, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/protocol.py#L1044-L1055" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/profiler/wrapper.py", "first": 308, "last": 320, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L308-L320" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1146, "last": 1153, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1146-L1153" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/entrypoints/serve/profile/api_router.py", "first": 21, "last": 34, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/profile/api_router.py#L21-L34" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/config/profiler.py", "first": 118, "last": 127, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/profiler.py#L118-L127" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/profiler/wrapper.py", "first": 215, "last": 232, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L215-L232" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "python/sglang/profiler.py", "first": 51, "last": 68, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/profiler.py#L51-L68" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 1168, "last": 1173, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L1168-L1173" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "python/sglang/profiler.py", "first": 73, "last": 137, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/profiler.py#L73-L137" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1000, "last": 1015, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1000-L1015" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "python/sglang/srt/utils/profile_utils.py", "first": 476, "last": 487, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/profile_utils.py#L476-L487" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/v1/worker/gpu_worker.py", "first": 1027, "last": 1041, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/worker/gpu_worker.py#L1027-L1041" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "python/sglang/srt/server_args.py", "first": 1941, "last": 1941, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1941-L1941" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "python/sglang/srt/server_args.py", "first": 8877, "last": 8882, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L8877-L8882" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "python/sglang/srt/server_args.py", "first": 1890, "last": 1897, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1890-L1897" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "vllm/profiler/wrapper.py", "first": 215, "last": 232, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L215-L232" }, { "page": "labs/10-profile-a-decode-step/README.html", "path": "wrapper.py", "first": 217, "last": 217, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/profiler/wrapper.py", "resolution_status": "same-section-qualified-anchor", "resolution_evidence": { "record_index": 3471, "page": "labs/10-profile-a-decode-step/README.html", "section": "exercises", "engine": "vllm", "path": "vllm/profiler/wrapper.py", "first": 215, "last": 232 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/profiler/wrapper.py#L217-L217" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 1778, "last": 1789, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1778-L1789" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 622, "last": 644, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L622-L644" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 730, "last": 732, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L730-L732" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 1432, "last": 1451, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1432-L1451" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/lib/endpoint_request_func.py", "first": 873, "last": 877, "engine": "vllm", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/lib/endpoint_request_func.py#L873-L877" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 1574, "last": 1578, "engine": "vllm", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1574-L1578" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 1714, "last": 1719, "engine": "sglang", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L1714-L1719" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "python/sglang/srt/entrypoints/openai/protocol.py", "first": 340, "last": 375, "engine": "sglang", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/protocol.py#L340-L375" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "python/sglang/srt/entrypoints/http_server.py", "first": 1843, "last": 1843, "engine": "sglang", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/http_server.py#L1843-L1843" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "python/sglang/benchmark/serving.py", "first": 1779, "last": 1780, "engine": "sglang", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L1779-L1780" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 1968, "last": 1969, "engine": "vllm", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1968-L1969" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 471, "last": 484, "engine": "vllm", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L471-L484" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 1666, "last": 1677, "engine": "vllm", "section": "one-harness", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1666-L1677" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 1692, "last": 1697, "engine": "vllm", "section": "protocol", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1692-L1697" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "python/sglang/benchmark/serving.py", "first": 2598, "last": 2603, "engine": "sglang", "section": "protocol", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L2598-L2603" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "python/sglang/benchmark/serving.py", "first": 2587, "last": 2591, "engine": "sglang", "section": "protocol", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/benchmark/serving.py#L2587-L2591" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 1678, "last": 1687, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L1678-L1687" }, { "page": "labs/11-end-to-end-benchmark/README.html", "path": "vllm/benchmarks/serve.py", "first": 963, "last": 973, "engine": "vllm", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/benchmarks/serve.py#L963-L973" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/logger.py", "first": 22, "last": 25, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/logger.py#L22-L25" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/utils/common.py", "first": 2338, "last": 2343, "engine": "sglang", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L2338-L2343" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/engine/arg_utils.py", "first": 2891, "last": 2899, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/engine/arg_utils.py#L2891-L2899" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/entrypoints/serve/utils/request_logger.py", "first": 44, "last": 68, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/utils/request_logger.py#L44-L68" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/server_args.py", "first": 1518, "last": 1528, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1518-L1528" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/utils/request_logger.py", "first": 107, "last": 111, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/request_logger.py#L107-L111" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/managers/schedule_batch.py", "first": 1803, "last": 1813, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_batch.py#L1803-L1813" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/managers/scheduler_components/output_streamer.py", "first": 184, "last": 190, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/scheduler_components/output_streamer.py#L184-L190" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/utils/common.py", "first": 2338, "last": 2339, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L2338-L2339" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/utils/common.py", "first": 2328, "last": 2337, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/common.py#L2328-L2337" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/logger.py", "first": 271, "last": 287, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/logger.py#L271-L287" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/config/vllm.py", "first": 796, "last": 812, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/config/vllm.py#L796-L812" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/logger.py", "first": 305, "last": 310, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/logger.py#L305-L310" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/multimodal_gen/runtime/utils/logging_utils.py", "first": 431, "last": 451, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/multimodal_gen/runtime/utils/logging_utils.py#L431-L451" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/multimodal_gen/envs.py", "first": 23, "last": 23, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/multimodal_gen/envs.py#L23-L23" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/v1/engine/core.py", "first": 1428, "last": 1430, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/engine/core.py#L1428-L1430" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/server_args.py", "first": 1642, "last": 1646, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1642-L1646" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "vllm/v1/metrics/loggers.py", "first": 310, "last": 313, "engine": "vllm", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/v1/metrics/loggers.py#L310-L313" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/managers/tokenizer_manager.py", "first": 1619, "last": 1624, "engine": "sglang", "section": "expect", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/tokenizer_manager.py#L1619-L1624" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "request_logger.py", "first": 44, "last": 68, "engine": null, "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "vllm/entrypoints/serve/utils/request_logger.py", "resolution_status": "same-page-qualified-anchor", "resolution_evidence": { "record_index": 3494, "page": "labs/12-read-a-request-through-the-code/README.html", "section": "run", "engine": "vllm", "path": "vllm/entrypoints/serve/utils/request_logger.py", "first": 44, "last": 68 }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/utils/request_logger.py#L44-L68" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "request_logger.py", "first": 21, "last": 33, "engine": null, "section": "exercises", "declared_engine": "vllm", "declared_path": "vllm/entrypoints/serve/utils/request_logger.py", "reviewed_symbol": "logger", "resolved_engine": "vllm", "resolved_path": "vllm/entrypoints/serve/utils/request_logger.py", "resolution_status": "reviewed-context-and-pinned-symbol", "resolution_evidence": { "engine": "vllm", "path": "vllm/entrypoints/serve/utils/request_logger.py", "symbol": "logger", "method": "Authored per-occurrence context qualification; symbol rechecked at cited pinned lines" }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/entrypoints/serve/utils/request_logger.py#L21-L33" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/utils/request_logger.py", "first": 198, "last": 211, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/utils/request_logger.py#L198-L211" }, { "page": "labs/12-read-a-request-through-the-code/README.html", "path": "python/sglang/srt/entrypoints/openai/serving_base.py", "first": 88, "last": 90, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/entrypoints/openai/serving_base.py#L88-L90" }, { "page": "labs/13-batch-invariance/README.html", "path": "tests/v1/determinism/utils.py", "first": 26, "last": 36, "engine": null, "section": null, "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "resolved_engine": "vllm", "resolved_path": "tests/v1/determinism/utils.py", "resolution_status": "unique-exact-path-in-pinned-trees", "resolution_evidence": { "path": "tests/v1/determinism/utils.py", "checked_engines": [ "sglang", "vllm" ], "engine": "vllm", "pins": { "vllm": "a556f3fccb701e5618d84d547ff454c56a1bfdfb", "sglang": "7d893255c359bb8ab74d2870c8ac865fb57230d6" } }, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/utils.py#L26-L36" }, { "page": "labs/13-batch-invariance/README.html", "path": "vllm/envs.py", "first": 627, "last": 629, "engine": "vllm", "section": null, "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/envs.py#L627-L629" }, { "page": "labs/13-batch-invariance/README.html", "path": "vllm/model_executor/layers/batch_invariant.py", "first": 916, "last": 929, "engine": "vllm", "section": null, "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/vllm/model_executor/layers/batch_invariant.py#L916-L929" }, { "page": "labs/13-batch-invariance/README.html", "path": "benchmarks/benchmark_batch_invariance.py", "first": 148, "last": 152, "engine": "vllm", "section": "what", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/benchmarks/benchmark_batch_invariance.py#L148-L152" }, { "page": "labs/13-batch-invariance/README.html", "path": "tests/v1/determinism/conftest.py", "first": 8, "last": 12, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/conftest.py#L8-L12" }, { "page": "labs/13-batch-invariance/README.html", "path": "tests/v1/determinism/test_batch_invariance.py", "first": 458, "last": 470, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/test_batch_invariance.py#L458-L470" }, { "page": "labs/13-batch-invariance/README.html", "path": "tests/v1/determinism/utils.py", "first": 39, "last": 40, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/utils.py#L39-L40" }, { "page": "labs/13-batch-invariance/README.html", "path": "python/sglang/test/test_deterministic.py", "first": 460, "last": 471, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/test/test_deterministic.py#L460-L471" }, { "page": "labs/13-batch-invariance/README.html", "path": "python/sglang/test/test_deterministic_utils.py", "first": 12, "last": 18, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/test/test_deterministic_utils.py#L12-L18" }, { "page": "labs/13-batch-invariance/README.html", "path": "python/sglang/srt/server_args.py", "first": 1906, "last": 1910, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/server_args.py#L1906-L1910" }, { "page": "labs/13-batch-invariance/README.html", "path": "python/sglang/srt/arg_groups/overrides.py", "first": 2097, "last": 2108, "engine": "sglang", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/arg_groups/overrides.py#L2097-L2108" }, { "page": "labs/13-batch-invariance/README.html", "path": "tests/v1/determinism/test_online_batch_invariance.py", "first": 144, "last": 161, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/test_online_batch_invariance.py#L144-L161" }, { "page": "labs/13-batch-invariance/README.html", "path": "tests/v1/determinism/test_online_batch_invariance.py", "first": 126, "last": 133, "engine": "vllm", "section": "run", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/vllm-project/vllm/blob/a556f3fccb701e5618d84d547ff454c56a1bfdfb/tests/v1/determinism/test_online_batch_invariance.py#L126-L133" }, { "page": "labs/13-batch-invariance/README.html", "path": "python/sglang/srt/managers/schedule_policy.py", "first": 1393, "last": 1402, "engine": "sglang", "section": "exercises", "declared_engine": null, "declared_path": null, "reviewed_symbol": null, "status": "range-exists", "url": "https://github.com/sgl-project/sglang/blob/7d893255c359bb8ab74d2870c8ac865fb57230d6/python/sglang/srt/managers/schedule_policy.py#L1393-L1402" } ] }