[Bug]: additional_drop_params not working in proxy mode

[uzhao]

What happened?

After I upgrade to 1.60, this param no longer working for me.
In my proxy config, I have

  - model_name: o3-mini
    litellm_params:
      model: github/o3-mini
      api_key: "minimini"
      max_completion_tokens: 100000
      reasoning_effort: "high"
      additional_drop_params: ["max_tokens", "temperature", "top_p"]
    model_info:
      id: github-o3-mini

litellm_settings:
  drop_params: true
  request_timeout: 300
  set_verbose: true

Then

curl http://127.0.0.1:4000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer bigbig" \
  -d '{
    "model": "o3-mini",
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant."
      },
      {
        "role": "user",
        "content": "Hello!"
      }
    ],
    "max_completion_tokens": 100000,
    "reasoning_effort": "high"
  }'

I got such a return
{"error":{"message":"litellm.BadRequestError: GithubException - 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'...

I tried downgrade to 1.59.10 but this still not solved. I remember I modified proxy_server.py in older version and manually removed these params but this no longer working

Relevant log output

Are you a ML Ops Team?

No

What LiteLLM version are you on ?

1.60

Twitter / LinkedIn details

No response