litellm - 💡(How to fix) Fix [Bug]: Claude Code + Microsoft Foundry (Azure) - Unsupported 'reasoning.effort' and AnthropicResponse Pydantic Validation Error [1 participants]

##Error 1 (Parameter):

litellm.exceptions.BadRequestError: litellm.BadRequestError: OpenAIException - { "error": { "message": "Unsupported parameter: 'reasoning.effort' is not supported with this model.", "type": "invalid_request_error", "param": "reasoning.effort", "code": "unsupported_parameter" } }

Error 2 (Pydantic):

Exception: [Non-Blocking] LiteLLM.Success_Call Error: 1 validation error for AnthropicResponse Input should be a valid dictionary or instance of AnthropicResponse [type=model_type, input_value=ResponseIncompleteEvent(...), input_type=ResponseIncompleteEvent] For further information visit https://errors.pydantic.dev/2.12/v/model_type

TL;DR

Adding additional_drop_params: ["reasoning.effort"] to the config.yaml file may resolve the Parameter Mapping Failure issue.

Guidance

• Verify that the additional_drop_params configuration option is correctly applied by checking the LiteLLM logs for the dropped parameter.
• Investigate the Pydantic ValidationError by reviewing the AnthropicResponse model definition and the ResponseIncompleteEvent or ResponseCompletedEvent data being validated against it.
• Consider updating the AnthropicResponse model to accommodate the ResponseIncompleteEvent or ResponseCompletedEvent data structure.
• Check the Microsoft Foundry documentation to confirm the supported parameters for the Kimi-K2.6 model.

Example

No code snippet is provided as the issue is more related to configuration and model compatibility.

Notes

The solution may not work if the additional_drop_params option does not correctly drop the reasoning.effort parameter or if the Pydantic ValidationError is caused by a more complex issue.

Recommendation

Apply workaround: Add additional_drop_params: ["reasoning.effort"] to the config.yaml file, as this is a targeted solution to the Parameter Mapping Failure issue and does not require updating the LiteLLM version.