sgl-kernel/python/sgl_kernel/sparse_flash_attn.py · ee71ed8a412709416dd688992dd084b1c0640495 · change / sglang · GitLab

Find file Blame History Permalink

[Feat] QWen-1M context support[1/2]: Update block sparse attention backend utils kernel (#5847) · ee71ed8a
PGFLMG authored Apr 29, 2025
```
Co-authored-by: sighingnow <sighingnow@gmail.com>
```
ee71ed8a

sparse_flash_attn.py 10.2 KB