LLM request failed with GPT 5 models

[alexpouliquen]

Description

Hello,

I am getting this error while trying to request gpt-5-mini, gpt-5-nano andgpt-5.

"error": {
    "message": "Unsupported value: 'temperature' does not support 0.5 with this model. Only the default (1) value is supported.",
    "type": "invalid_request_error",
    "param": "temperature",
    "code": "unsupported_value"
}

Expected Behavior

Expect the LLM request to not failed. I guess by setting temperature: 1 or just not passing it.

Steps to Reproduce

1. Run openfang start
2. Go on web dashboard > chat
3. Choose one of the mentioned model
4. Send a message

OpenFang Version

0.3.29

Operating System

macOS (Apple Silicon)

Logs / Screenshots

No response

[jaberjaber23]

Investigated thoroughly. GPT-5, o3, o4-mini are all in the model catalog with proper pricing and driver support. The OpenAI driver has adaptive handling for max_completion_tokens vs max_tokens and temperature rejection for o-series models. This is most likely an API key configuration issue. Make sure OPENAI_API_KEY is set and that your OpenAI account has access to the model you're trying to use.