[Bug]: Cannot see the http request that was sent

[mrm1001]

What happened?

This is related to: #489.

I'm trying to see the raw request that was sent to the LLM endpoint. When I set litellm.set_verbose=True, I only get to see the raw response. I've also tried setting os.environ['LITELLM_LOG'] = 'DEBUG'.

My code:

import litellm

response = litellm.completion(
    model="gemini/gemini-2.0-flash-exp", 
    messages=[{"role": "user", "content": "hi there"}],    
) 

The message that gets printed in stdout:

�[92m11:02:02 - LiteLLM:WARNING�[0m: utils.py:322 - `litellm.set_verbose` is deprecated. Please set `os.environ['LITELLM_LOG'] = 'DEBUG'` for debug logs.
SYNC kwargs[caching]: False; litellm.cache: None; kwargs.get('cache')['no-cache']: False
Final returned optional params: {}
RAW RESPONSE:
{
  "candidates": [
    {
      "content": {
        "parts": [
          {
            "text": "Hi! How can I help you today?\n"
          }
        ],
        "role": "model"
      },
      "finishReason": "STOP",
      "safetyRatings": [
        {
          "category": "HARM_CATEGORY_HATE_SPEECH",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_HARASSMENT",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
          "probability": "NEGLIGIBLE"
        }
      ],
      "avgLogprobs": -0.12977735996246337
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 3,
    "candidatesTokenCount": 10,
    "totalTokenCount": 13
  },
  "modelVersion": "gemini-2.0-flash-exp"
}

Relevant log output

Are you a ML Ops Team?

No

What LiteLLM version are you on ?

v1.58.2

Twitter / LinkedIn details

No response

[superpoussin22]

if you want to see prompts and response you could use langfuse

[mrm1001]

Langfuse is great, but this is very basic functionality that the library should provide. I have been trying to see the raw request for hours for another issue I'm experiencing (#7804) and have not managed!

Also, the docs are wrong.