[bug]: use session tracking using langfuse in litellm sdk

[mubashir1osmani]

The Feature

sending the "metadata" payload works with litellm proxy but the sdk ignores it. the goal is to be able to pass session id in the payload for OpenAI api calls

litellm.api_key = "xxx"
litellm._turn_on_debug()

response = litellm.completion(
    model="gpt-4.1-nano",  # Change this to the model deployed on your LiteLLM server
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Write a haiku about the ocean burning"},
    ],
    metadata={"session_id": "1234"}
)

print(response.choices[0].message["content"])```

### Motivation, pitch

Ensure metadata is available in logging for all integrations

### LiteLLM is hiring a founding backend engineer, are you interested in joining us and shipping to all our users?

No

### Twitter / LinkedIn details

_No response_

[NickBak08]

For me the session_id works in langfuse with litellm 1.75.9 and by using litellm.callbacks = ["langfuse_otel"]. I am not using openai models but an ollama model.

[emerzon]

For me the session_id works in langfuse with litellm 1.75.9 and by using litellm.callbacks = ["langfuse_otel"]. I am not using openai models but an ollama model.

This issue is specific to the SDK client, not the proxy

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.