Skip to content
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

[V1] Fix when max_model_len is not divisible by block_size #10903

Merged
merged 3 commits into from
Dec 5, 2024

Conversation

WoosukKwon
Copy link
Collaborator

@WoosukKwon WoosukKwon commented Dec 4, 2024

Fixes #10891

Thanks @ywang96 for investigating the bug and pinpointing the source of the bug.

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Copy link

github-actions bot commented Dec 4, 2024

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

  • Add ready label to the PR
  • Enable auto-merge.

🚀

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Copy link
Member

@ywang96 ywang96 left a comment

Choose a reason for hiding this comment

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

I've tested this PR for both text models and image models and the change looks good to me!

As discussed offline, we should add some unit tests for V1 engine around issue like this, since the bug only occur in certain situations without an explicit error.

@ywang96 ywang96 added the ready ONLY add when PR is ready to merge/full CI is needed label Dec 5, 2024
@WoosukKwon WoosukKwon merged commit 2a56e12 into main Dec 5, 2024
53 of 59 checks passed
@WoosukKwon WoosukKwon deleted the v1-max-model-len branch December 5, 2024 00:54
sleepwalker2017 pushed a commit to sleepwalker2017/vllm that referenced this pull request Dec 13, 2024
…ect#10903)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
BKitor pushed a commit to BKitor/vllm that referenced this pull request Dec 30, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ready ONLY add when PR is ready to merge/full CI is needed
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[Bug, V1]: LlaVa outputs wrong results in batch inference with V1 code(V0 code is correct)
2 participants