Xin Yang d08b356ee0 [Perf] Create TMA-aligned input scale tensor for DeepGemm on Hopper (#32619) ...

Signed-off-by: Xin Yang <xyangx@amazon.com>

2026-01-22 15:47:04 -05:00

..

__init__.py

[Chore][1/2] Drop v0.14 deprecations (#31285)

2025-12-24 09:54:01 -08:00

argparse_utils.py

[Docs] Improve malformed exception caused by backslash line continuations (#31694)

2026-01-05 17:51:54 -08:00

async_utils.py

[mypy] Pass type checking for vllm/utils and vllm/v1/pool (#29666)

2025-11-28 20:40:47 +08:00

cache.py

[Docs] Mock all imports for docs (#27873)

2025-11-01 10:02:23 +00:00

collection_utils.py

[Model] Extend collect_children and no_init_weights contexts (#32757)

2026-01-22 08:20:27 +00:00

counter.py

[Misc] Separate out utils.counter and move utils.Device to engine (#27588)

2025-10-28 12:20:46 +00:00

deep_gemm.py

[Perf] Create TMA-aligned input scale tensor for DeepGemm on Hopper (#32619)

2026-01-22 15:47:04 -05:00

flashinfer.py

[Kernel][Performance] Enable smaller Scaling Factor tiling for NVFP4 small-batch decoding (#30885)

2026-01-13 15:22:53 -08:00

func_utils.py

[Misc] Move utils to avoid conflicts with stdlib, and move tests (#27169)

2025-10-19 05:20:55 -07:00

gc_utils.py

[Doc]: fixing typos in multiple files. (#29685)

2025-11-28 08:41:41 -08:00

hashing.py

[Core] Add xxHash as a high-performance hash option for accelerating prefix caching (#29163)

2025-12-03 16:06:57 +00:00

import_utils.py

[AMD][ROCm] MoRI EP: a high-performance all2all backend (#28664)

2026-01-22 16:33:18 +08:00

jsontree.py

[mypy] Pass type checking for vllm/utils and vllm/v1/pool (#29666)

2025-11-28 20:40:47 +08:00

math_utils.py

[Performance] Improve Triton prefill attention kernel's performance (#32403)

2026-01-17 20:19:59 -08:00

mem_constants.py

[Chore] Separate out vllm.utils.mem_utils (#27143)

2025-10-18 10:06:59 +00:00

mem_utils.py

[Misc] Make mem utils can be reused by other platforms (#32322)

2026-01-14 03:46:01 -08:00

nccl.py

[mypy] Pass type checking for vllm/utils and vllm/v1/pool (#29666)

2025-11-28 20:40:47 +08:00

network_utils.py

[Misc] Replace urllib's urlparse with urllib3's parse_url (#32746)

2026-01-22 16:37:15 +08:00

nvtx_pytorch_hooks.py

[Feature] Add Layer-wise NVTX Support (#29990)

2025-12-05 11:20:07 +00:00

platform_utils.py

[BugFix][ROCm] Fix get_cu_count missing variable error (#28608)

2025-11-13 05:18:56 +00:00

profiling.py

[Chore] Separate out profiling utilities from vllm.utils (#27150)

2025-10-18 19:12:01 +00:00

registry.py

[mypy] Pass type checking for vllm/utils and vllm/v1/pool (#29666)

2025-11-28 20:40:47 +08:00

serial_utils.py

[Frontend] Binary embedding response does not return metadata by setting encoding_format to bytes_only. (#30249)

2025-12-08 12:01:21 +00:00

system_utils.py

[Bugfix] Force using spawn multiprocess method when it's the WSL platform (#32749)

2026-01-21 09:35:55 +00:00

tensor_schema.py

Update Optional[x] -> x | None and Union[x, y] to x | y (#26633)

2025-10-12 09:51:31 -07:00

torch_utils.py

[misc] Remove is_torch_equal_or_newer(2.4) cases (#32296)

2026-01-13 23:22:07 -08:00