👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

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 either: Add ready label to the PR or enable auto-merge.

🚀

Who is the log message for though? The person setting the request-level backend is going to see the exception, either in a Python program using vllm as a library, or in the HTTP response from the API.

My thinking was:

• A library user who sets the backend in the request will see the warning telling them to not do that
• A library user who sets the backend to something different to the engine backend will see the warning and the error
• A HTTP user who sets the backend to something different to the engine backend will see the error

Who is the log message for though? The person setting the request-level backend is going to see the exception, either in a Python program using vllm as a library, or in the HTTP response from the API.

My thinking was:

• A library user who sets the backend in the request will see the warning telling them to not do that
• A library user who sets the backend to something different to the engine backend will see the warning and the error
• A HTTP user who sets the backend to something different to the engine backend will see the error

Won't all of these cases see the exception (either as the exception, or the HTTP error response) ?

Won't all of these cases see the exception (either as the exception, or the HTTP error response) ?

I thought that if a user set the request structured output backend to the same as the engine structured output backend, they wouldn't see the error?

edit: from offline discussion we decided that the warning only helps offline users and makes server logs super noisy when the problem doesn't concern the server admin. I've removed the warning.