Skip to content

Conversation

@fruworg
Copy link

@fruworg fruworg commented Sep 29, 2025

Please describe the purpose of this pull request.
Fix for #2681 (comment)

How to test
docker compose up -d
curl -H http://localhost:8283/v1/models/ -s | jq

  {
    "model": "<vllm-model>",
    "model_endpoint_type": "openai",
    "model_endpoint": "http://vllm:8000/v1",
    "provider_name": "vllm",
    "provider_category": "base",
    "model_wrapper": "chatml",
    "context_window": 2048,
    "put_inner_thoughts_in_kwargs": true,
    "handle": "vllm/<vllm-model>",
    "temperature": 0.7,
    "max_tokens": null,
    "enable_reasoner": true,
    "reasoning_effort": null,
    "max_reasoning_tokens": 0,
    "frequency_penalty": null,
    "compatibility_type": null,
    "verbosity": null,
    "tier": null
  }

Have you tested this PR?
Yup.

Related issues or PRs
#2681 (comment)

Is your PR over 500 lines of code?
Nope.

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.

5 participants