[Bug]: Incorrect token context for codestral/codestral-latest

[FractalWire]

What happened?

Hello,

I noticed that the current context window for codestral is set to 32000 intput (https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json#L2205)
I think this is incorrect as stated here: https://mistral.ai/en/news/codestral-2501

It should be now 256k context window.
I didn't do a PR because I'm not quite sure what to do with the other parameters max-tokens and max-input-token

Relevant log output

Are you a ML Ops Team?

No

What LiteLLM version are you on ?

Aider v0.74.1

Twitter / LinkedIn details

No response

[miraclebakelaser]

{
      "id": "codestral-latest",
      "object": "model",
      "created": 1739046941,
      "owned_by": "mistralai",
      "capabilities": {
        "completion_chat": true,
        "completion_fim": true,
        "function_calling": true,
        "fine_tuning": false,
        "vision": false
      },
      "name": "codestral-2501",
      "description": "Official codestral-2501 Mistral AI model",
      "max_context_length": 262144,
      "aliases": [
        "codestral-2501",
        "codestral-2412",
        "codestral-2411-rc5"
      ],
      "deprecation": null,
      "default_model_temperature": 0.3,
      "type": "base"
    },

https://api.mistral.ai/v1/models/{model_id}:
max_context_length is 262144. max-tokens and max-input tokens are the same.

From the documentation:

The maximum number of tokens to generate in the completion. The token count of your prompt plus max_tokens cannot exceed the model's context length.

Thus, both max_input_tokens and max_output_tokens should be 262144.