Skip to content

Support cascade-attention #921

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.

Sign up for GitHub

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

Jump to bottom
Draft
wants to merge 7 commits into
base: master
Choose a base branch
from
Draft

Support cascade-attention #921

wants to merge 7 commits into from

Conversation

0x45f
Copy link
Collaborator

@0x45f 0x45f commented Aug 25, 2025 •
edited
Loading

PR Category

Operator

Type of Change

Refactor

Description

Support cascade-attention
https://github.com/vllm-project/vllm/blob/v0.8.5/tests/kernels/attention/test_cascade_flash_attn.py
https://github.com/vllm-project/vllm/blob/v0.8.5/vllm/v1/attention/backends/flash_attn.py#L727

Issue

Progress

  • Change is properly reviewed (1 reviewer required, 2 recommended).
  • Change is responded to an issue.
  • Change is fully covered by a UT.

Performance

tongxin
tongxin reviewed Aug 26, 2025
View reviewed changes
Copy link
Contributor

@tongxin tongxin left a comment

Choose a reason for hiding this comment

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

Great, benchmark passed! Next we should add more correctness tests for this patch.

@tongxin
Copy link
Contributor

tongxin commented Sep 8, 2025

The lse store mask error was the cause of the failing tests and it's fixed now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@tongxin tongxin tongxin left review comments

At least 1 approving review is required to merge this pull request.

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@0x45f @tongxin