Skip to content

Conversation

joerunde
Copy link
Collaborator

Description

It turns out when prompt_logprobs are enabled, the static batching model runner will crash when calculating logprobs.

This draft has a failing test, should be a simple fix

Copy link

👋 Hi! Thank you for contributing to vLLM support on Spyre.
Just a reminder: Make sure that your code passes all the linting checks, otherwise your PR won't be able to be merged. To do so, first install the linting requirements, then run format.sh and commit the changes. This can be done with uv directly:

uv sync --frozen --group lint --active --inexact

Or this can be done with pip:

uv pip compile --group lint > requirements-lint.txt
pip install -r requirements-lint.txt
bash format.sh

Now you are good to go 🚀


for r in responses:
assert r.outputs[0].logprobs is not None
force_engine_shutdown(llm)
Copy link
Collaborator

@prashantgupta24 prashantgupta24 Aug 1, 2025

Choose a reason for hiding this comment

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

can we add some comments on when we have to use force_engine_shutdown? Also I wonder if we can make it such that it's part of pytest teardown automatically in case someone forgets to add this which would result in the weird addr-in-use error

Copy link
Collaborator

Choose a reason for hiding this comment

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

Yeah, maybe we could make it an auto-use fixture.

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants