Set Flux models to eval mode to restore FP8 performance #10330
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Fixes performance regression introduced in comfyanonymous#9854 where the fp8_linear optimization path was being skipped during inference because Flux models were not explicitly set to eval mode.
|
Thanks for reporting this, #10331 will fix it on all models not just on flux. |
|
Thanks for the quick merge. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes performance regression introduced in #9854 where the fp8_linear optimization path was being skipped during inference because Flux models were not explicitly set to eval mode.
This increased my Flux workflow's it/s on a 5090 back to 2.75 it/s, up from 1.85 it/s which I was getting on latest master.
Specifically, 7be2b49 was the first commit chronologically where my it/s dropped.