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

[Bugfix][Model] Jamba assertions and no chunked prefill by default for Jamba #6784

Merged
merged 2 commits into from
Jul 27, 2024

Conversation

tomeras91
Copy link
Contributor

#6666 enabled chunked prefill by default for long-context (>32K) models. Jamba is such a model, but it doesn't support chunked prefill.

This PR:

  1. Doesn't enable chunked prefill by default for Jamba (currently the only model with seqlen agnostic layers)
  2. Adds assertions inside Jamba code that make sure it's not used with chunked prefill or prefix caching which are currently not supported

Copy link

👋 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 consists a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of default ones by unblocking the steps in your fast-check build on Buildkite UI.

Once the PR is approved and ready to go, please make sure to run full CI as it is required to merge (or just use auto-merge).

To run full CI, you can do one of these:

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

🚀

@tomeras91
Copy link
Contributor Author

/ready

@github-actions github-actions bot added the ready ONLY add when PR is ready to merge/full CI is needed label Jul 25, 2024
@WoosukKwon WoosukKwon self-assigned this Jul 27, 2024
Copy link
Collaborator

@WoosukKwon WoosukKwon left a comment

Choose a reason for hiding this comment

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

LGTM! Sorry for introducing the bug to the Jamba model support.

@WoosukKwon WoosukKwon merged commit ed94e4f into vllm-project:main Jul 27, 2024
78 of 85 checks passed
@tomeras91 tomeras91 deleted the jamba-assertions branch August 12, 2024 15:00
kylesayrs pushed a commit to neuralmagic/vllm that referenced this pull request Aug 17, 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.

2 participants