Skip to content

Page Attention GPU Kernel Integration #1137

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 4 commits into
base: main
Choose a base branch
from

Conversation

christ-tt
Copy link
Contributor

Features:

  • Add paged attention as a valid backend for decoding.
  • Integrate Paged Attention GPU Kernel from Jax's implementation with customized bias and mask.

Testing:

  • Updated pallas kernel is tested in common/flash_attention/decoding_test.py
  • Updated flash_attention_implementation wrapper is tested in common/flash_attention/utils_test.py

TODO:

  • Update FlashAttention Layer when we have PagedKVCache implementation integrated.

@christ-tt christ-tt requested review from ruomingp, markblee and a team as code owners April 28, 2025 19:01
Copy link
Contributor

@markblee markblee left a comment

Choose a reason for hiding this comment

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

Missing internal review.

@christ-tt christ-tt requested a review from markblee May 2, 2025 23:41
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.

2 participants