Ming YangandClaude 9b4bb415dd [cuda-graph] Size breakable-graph shared buffer from warmup output; slice by produced row count (#30834) ...

Co-authored-by: Claude <noreply@anthropic.com>

2026-07-11 10:41:44 -07:00

..

__init__.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

base_cuda_graph_backend.py

Fix circular import when sglang.srt.model_executor.runner_backend is imported first (#28002)

2026-06-16 12:04:34 -07:00

breakable_cuda_graph_backend.py

[cuda-graph] Size breakable-graph shared buffer from warmup output; slice by produced row count (#30834)

2026-07-11 10:41:44 -07:00

cuda_graph_dedup_mixin.py

CUDA graph executable dedup via cudaGraphExecUpdate (#29625)

2026-06-29 13:56:33 -07:00

full_cuda_graph_backend.py

[Refactor] Share CUDA graph memory pool across prefill and decode (#28973)

2026-06-24 10:18:20 -07:00

tc_piecewise_cuda_graph_backend.py

[refactor] Adopt get_parallel() everywhere and close out the parallel wrapper surface (#30492)

2026-07-09 02:09:39 -07:00

utils.py

[Experimental] Full Cuda Graph Support for Prefill (#27988)

2026-07-06 18:13:03 -07:00