### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

This issue is closely related to the [issue #9583](https://github.com/BerriAI/litellm/issues/9583).

Description: Systematic error encountered when calling mistral-large-3 hosted on Microsoft Foundry with the parameter `max_completion_tokens`

```json { "error": { "message": "litellm.APIConnectionError: APIConnectionError: Azure_aiException - Extra data: line 2 column 1 (char 209). Received Model Group=mistral-large-3\nAvailable Model Group Fallbacks=None", "type": null, "param": null, "code": "500" } } ```

This error is only encountered with this model. When switching to any other model the following requests works flawlessly:

```bash curl --location 'http://localhost:4000/v1/chat/completions' --header 'Content-Type: application/json' --header 'Authorization: Bearer sk-1234' --data '{ "model": "mistral-large-3", "messages": [ { "role": "system", "content": [ { "type": "text", "text": "Faux contexte" } …