Skip to content

[CUDA] Update Flash Attention to support head_sink for smooth softmax in GQA#25358

Closed
tianleiwu wants to merge 2 commits into
mainfrom
tlwu/gqa_head_sink_cuda
Closed

[CUDA] Update Flash Attention to support head_sink for smooth softmax in GQA#25358
tianleiwu wants to merge 2 commits into
mainfrom
tlwu/gqa_head_sink_cuda

add test

f8cd59e
Select commit
Loading
Failed to load commit list.
GitHub Advanced Security / lintrunner succeeded Jul 10, 2025 in 4s

1 new alert

New alerts in code changed by this pull request

  • 1 note

See annotations below for details.

View all branch alerts.

Annotations

Check notice on line 180 in onnxruntime/test/python/transformers/test_gqa.py

See this annotation in the file changed.

Code scanning / lintrunner

RUFF/PLC0415 Note test

import should be at the top-level of a file.
See https://docs.astral.sh/ruff/rules/import-outside-top-level