microsoft / BitBLAS

BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.
MIT License
423 stars 34 forks source link

[TL] initial implement flashattention op in TL #202

Closed tzj-fxz closed 1 month ago

tzj-fxz commented 1 month ago
LeiWang1999 commented 1 month ago

thanks @tzj-fxz ,overall lgtm, should fix the test.

tzj-fxz commented 1 month ago

thanks @tzj-fxz ,overall lgtm, should fix the test.

OK, now CI tests are passed in the newest commit.

LeiWang1999 commented 1 month ago

Merged, thanks @tzj-fxz