AMD FA performance
#1714
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
{{ message }}
Uh oh!
There was an error while loading. Please reload this page.
-
For the best performance on AMD GPUs, we should use [https://github.com/Dao-AILab/flash-attention/tree/main/flash_attn/flash_attn_triton_amd] correct? And even though this path is named flash_attn_triton_amd, this is used for running ck FA as well?
Beta Was this translation helpful? Give feedback.
All reactions