Merged
Conversation
MeinhardZhou
pushed a commit
to bytedance-iaas/xDiT
that referenced
this pull request
Jun 24, 2025
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
FA3 now supports FP8 (torch.float8_e4m3fn) reasoning. However, the forward stage still lacks FP8 support.
I added the necessary parameters qkv scale for quantization and selected the Attention that supports FP8 for forwarding according to AttnType in yunchang (USP). This function will support USP forward of FP8 quantization.
In addition, pull requests have been initiated for quantization-related parameter support for yunchang and FA3.
yunchang (USP): https://github.com/feifeibear/long-context-attention/pull/151（merged into main）
FA3: Dao-AILab/flash-attention#1686