Skip to content

Conversation

@MondayCha
Copy link
Contributor

@MondayCha MondayCha commented Oct 27, 2025

This pull request enhances the vLLM inference simulator by implementing full support for the tool_choice option, specifically adding the capability to force the model to use a specific tool. This update brings the simulator's functionality into closer alignment with the official OpenAI API, allowing for more precise and sophisticated testing of tool-use scenarios.

Limitations: Currently, this PR does not handle the case where the tool type is custom, nor does it address the allowed_tools option. Adding custom-type tools will require refactoring more fields. For OpenAI's documentation on these fields, refer to tool-choice.

Fixes #235

@MondayCha MondayCha force-pushed the main branch 2 times, most recently from 2bc6b38 to 6e6d964 Compare October 27, 2025 12:49
@MondayCha MondayCha marked this pull request as draft October 27, 2025 12:59
@MondayCha MondayCha marked this pull request as ready for review October 27, 2025 16:01
@irar2
Copy link
Collaborator

irar2 commented Oct 28, 2025

@MondayCha Is the PR ready for review?

@MondayCha
Copy link
Contributor Author

@irar2 Yes, I've tested it normally.

I feel like custom can be put in the next PR for processing

@irar2
Copy link
Collaborator

irar2 commented Oct 28, 2025

/lgtm
/approve

@github-actions github-actions bot added the lgtm label Oct 28, 2025
@github-actions github-actions bot merged commit cd611e8 into llm-d:main Oct 28, 2025
4 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Configure the tool_choice option to use a specific tool

2 participants