Skip to content

If you use the OpenAI provider in Goose and use the local vLLM Server, a strange error will be output. #8581

Description

@ytooyama

Describe the bug
A clear and concise description of what the bug is.


To Reproduce
Steps to reproduce the behavior:

  • Use vLLM to launch an OpenAI API-compatible LLM Server
uv run python -m vllm.entrypoints.openai.api_server \
    --model google/gemma-4-26b-a4b-it \
    --dtype bfloat16 \
    --max-model-len 32768 \
    --gpu-memory-utilization 0.90 \
    --port 8080 \
    --host 0.0.0.0 \
    --enable-auto-tool-choice \
    --tool-call-parser gemma4
  • Set up an OpenAI provider with Goose

    • API Key:
    • API Host: http://(IP address of vLLM Server):8080
    • Base Path(OPENAI_BASE_PATH): v1/chat/completions
  • Select the model on vLLM Server in Goose (google/gemma-4-26b-a4b-it)

  • When you throw a request, the request is processed and the result is output to Goose.

Image
  • However, the following errors will be logged..

Expected behavior
No error is displayed.


Screenshots

  • "POST /v1/chat/completions HTTP/1.1" 200 OK is output in the vLLM log, but "gpt-4o-mini does not exist." There is also an error.
(APIServer pid=39819) ERROR 04-16 14:34:18 [serving.py:193] Error with model error=ErrorInfo(message='The model `gpt-4o-mini` does not exist.', type='NotFoundError', param='model', code=404)

The entire error log:

(APIServer pid=39819) ERROR 04-16 14:34:10 [serving.py:193] Error with model error=ErrorInfo(message='The model `gpt-4o-mini` does not exist.', type='NotFoundError', param='model', code=404)
(APIServer pid=39819) INFO:     172.16.0.101:58385 - "POST /v1/chat/completions HTTP/1.1" 404 Not Found
(APIServer pid=39819) INFO:     172.16.0.101:58386 - "POST /v1/chat/completions HTTP/1.1" 200 OK
(APIServer pid=39819) ERROR 04-16 14:34:11 [serving.py:193] Error with model error=ErrorInfo(message='The model `gpt-4o-mini` does not exist.', type='NotFoundError', param='model', code=404)
(APIServer pid=39819) INFO:     172.16.0.101:58385 - "POST /v1/chat/completions HTTP/1.1" 404 Not Found
(APIServer pid=39819) ERROR 04-16 14:34:13 [serving.py:193] Error with model error=ErrorInfo(message='The model `gpt-4o-mini` does not exist.', type='NotFoundError', param='model', code=404)
(APIServer pid=39819) INFO:     172.16.0.101:58385 - "POST /v1/chat/completions HTTP/1.1" 404 Not Found
(APIServer pid=39819) INFO 04-16 14:34:14 [loggers.py:259] Engine 000: Avg prompt throughput: 335.7 tokens/s, Avg generation throughput: 35.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 1.3%, Prefix cache hit rate: 0.0%
(APIServer pid=39819) ERROR 04-16 14:34:18 [serving.py:193] Error with model error=ErrorInfo(message='The model `gpt-4o-mini` does not exist.', type='NotFoundError', param='model', code=404)
(APIServer pid=39819) INFO:     172.16.0.101:58386 - "POST /v1/chat/completions HTTP/1.1" 404 Not Found
(APIServer pid=39819) INFO:     172.16.0.101:58386 - "POST /v1/chat/completions HTTP/1.1" 200 OK
(APIServer pid=39819) INFO 04-16 14:34:24 [loggers.py:259] Engine 000: Avg prompt throughput: 15.0 tokens/s, Avg generation throughput: 5.2 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%, Prefix cache hit rate: 0.0%
(APIServer pid=39819) INFO 04-16 14:34:34 [loggers.py:259] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 0.0 tokens/s, Running: 0 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.0%, Prefix cache hit rate: 0.0%
(APIServer pid=39819) INFO:     172.16.0.101:58395 - "GET /v1/models HTTP/1.1" 200 OK

Please provide the following information

  • vLLM Server
    • OS: Red Hat Enterprise Linux 10.1
    • vLLM v0.19.0
  • Goose
    • OS: macOS Tahoe v26.4.1
    • Interface: UI
    • Version: v1.29.1

Additional context
vLLM Server is located in the company's environment and is accessed over VPN.
But it's working well.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions