OpenShift Lightspeed is using the deprecated max_tokens which will lead to the following error 
[ols.app.endpoints.health:health.py:59] ERROR: LLM connection check failed with - Error code: 400 - {'error': {'message': "Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.", 'type': 'invalid_request_error', 'param': 'max_tokens', 'code': 'unsupported_parameter'}}

    OpenShift Lightspeed 1.0
Model: gpt-5-mini

    1. 
    2.
    3.
    

    Using the models from above will lead to 400 errors

    Can use models from the above section

    - The max_tokens parameter is deprecated and max_completion_tokens shall be used instead according to https://platform.openai.com/docs/api-reference/chat/create.
- The AzureOpenAI code still has max_tokens in use https://github.com/openshift/lightspeed-service/blob/main/ols/src/llms/providers/azure_openai.py#L74