Replies: 1 comment
-
Any idea why xformers & FlashAttention cause such a speed degradation? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Settings:
Card: 3080Ti 12288MiB VRAM
cached latents of 7 samples, batch size 1
Results:
As far as I know xformers integrated FlashAttention.
I would recommand xformers for 12GB because you can use larger batch size.
Beta Was this translation helpful? Give feedback.
All reactions