Skip to content

Conversation

@leisuzz
Copy link
Contributor

@leisuzz leisuzz commented Aug 30, 2025

What does this PR do?

NPU FA is more suitable for the shape of BNSD. After converted the shape of q, k, and v, the performance will improve a lot.

Before submitting

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@leisuzz
Copy link
Contributor Author

leisuzz commented Aug 30, 2025

Cc: @sayakpaul @a-r-r-o-w , Please take a look at this PR, thanks for your help :)

Copy link
Contributor

@a-r-r-o-w a-r-r-o-w left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thanks!

@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@a-r-r-o-w a-r-r-o-w merged commit 827fad6 into huggingface:main Aug 30, 2025
10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants