[Kernel] Use flash-attn for decoding#3648
Merged
WoosukKwon merged 98 commits intovllm-project:mainfrom skrider:flash-attention-decodeMay 13, 2024
+313-65
Commits
Commits on Mar 27, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on Mar 28, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on Apr 22, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on May 6, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on May 7, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on May 8, 2024
Commits on May 9, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on May 10, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on May 11, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
Commits on May 13, 2024
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed