Wentao Ye 5c3fbfe46b

[Feature] Full Cuda Graph Support for Cutlass MLA and 6% E2E Throughput Improvement (#22763) ...

Signed-off-by: yewentao256 <zhyanwentao@126.com>

2025-08-15 06:27:30 +00:00

..

piecewise

[Feature] Full Cuda Graph Support for Cutlass MLA and 6% E2E Throughput Improvement (#22763)

2025-08-15 06:27:30 +00:00

__init__.py

[torch.compile] register allreduce operations as custom ops (#8526)

2024-09-16 22:57:57 -07:00

backend.py

[torch.compile][ROCm] Fuse quantization onto attention using a torch.compile pass (#16756)

2025-06-12 08:31:04 -07:00

test_async_tp.py

[Feature] Add async tensor parallelism for scaled mm (#20155)

2025-07-30 17:23:41 -04:00

test_basic_correctness.py

[Deprecation][2/N] Replace --task with --runner and --convert (#21470)

2025-07-27 19:42:40 -07:00

test_config.py

[Bugfix] VLLM_V1 supports passing other compilation levels (#19340)

2025-07-29 09:35:58 -04:00

test_full_graph.py

[Bugfix] Upgrade depyf to 0.19 and streamline custom pass logging (#20777)

2025-07-11 00:15:22 -07:00

test_functionalization.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

test_fusion_all_reduce.py

[CI] Initial tests for SM100 Blackwell runner (#21877)

2025-08-01 16:18:38 -07:00

test_fusion_attn.py

[Perf][fp8] Use CustomOp abstraction for fp8 quant for better perf (#19830)

2025-07-11 04:56:28 +00:00

test_fusion.py

[Perf][fp8] Use CustomOp abstraction for fp8 quant for better perf (#19830)

2025-07-11 04:56:28 +00:00

test_pass_manager.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

test_sequence_parallelism.py

[Deprecation][2/N] Replace --task with --runner and --convert (#21470)

2025-07-27 19:42:40 -07:00

test_silu_mul_quant_fusion.py

[Perf][fp8] Use CustomOp abstraction for fp8 quant for better perf (#19830)

2025-07-11 04:56:28 +00:00

test_wrapper.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00