Skip to content

Conversation

@DarkLight1337
Copy link
Member

@github-actions
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 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.

🚀

@mergify mergify bot added the documentation Improvements or additions to documentation label Nov 25, 2024
@DarkLight1337
Copy link
Member Author

Let me verify that the test passes first before adding ready label.

@DarkLight1337
Copy link
Member Author

Oh, this is awkward. By default HF assigns is_decoder=False for all models, even the text generation models.

@DarkLight1337
Copy link
Member Author

Maybe I have to use another config field for this...

Signed-off-by: DarkLight1337 <[email protected]>
@DarkLight1337 DarkLight1337 changed the title [Model] Support is_decoder HF config for Qwen2 model [Model] Support is_causal HF config field for Qwen2 model Nov 25, 2024
Signed-off-by: DarkLight1337 <[email protected]>
@DarkLight1337 DarkLight1337 added the ready ONLY add when PR is ready to merge/full CI is needed label Nov 25, 2024
@DarkLight1337 DarkLight1337 enabled auto-merge (squash) November 25, 2024 07:50
@DarkLight1337
Copy link
Member Author

Using is_causal passes the tests locally.

Signed-off-by: DarkLight1337 <[email protected]>
@DarkLight1337 DarkLight1337 merged commit ed46f14 into vllm-project:main Nov 25, 2024
51 checks passed
@DarkLight1337 DarkLight1337 deleted the qwen2-is-decoder branch November 25, 2024 10:22
afeldman-nm pushed a commit to neuralmagic/vllm that referenced this pull request Dec 2, 2024
sleepwalker2017 pushed a commit to sleepwalker2017/vllm that referenced this pull request Dec 13, 2024
anko-intel pushed a commit to HabanaAI/vllm-fork that referenced this pull request Feb 12, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

documentation Improvements or additions to documentation 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