litellm/cookbook/logging_observability/LiteLLM_Langfuse.ipynb

Use LiteLLM with Langfuse

https://docs.litellm.ai/docs/observability/langfuse_integration

Install Dependencies

In [ ]:

!pip install litellm langfuse

Set Env Variables

In [10]:

import litellm
from litellm import completion
import os

# from https://cloud.langfuse.com/
os.environ["LANGFUSE_PUBLIC_KEY"] = ""
os.environ["LANGFUSE_SECRET_KEY"] = ""


# OpenAI and Cohere keys
# You can use any of the litellm supported providers: https://docs.litellm.ai/docs/providers
os.environ['OPENAI_API_KEY']=""
os.environ['COHERE_API_KEY']=""

Set LangFuse as a callback for sending data

OpenAI completion call

In [8]:

# set langfuse as a callback, litellm will send the data to langfuse
litellm.success_callback = ["langfuse"]

# openai call
response = completion(
  model="gpt-3.5-turbo",
  messages=[
    {"role": "user", "content": "Hi 👋 - i'm openai"}
  ]
)

print(response)

{
  "id": "chatcmpl-85nP4xHdAP3jAcGneIguWATS9qdoO",
  "object": "chat.completion",
  "created": 1696392238,
  "model": "gpt-3.5-turbo-0613",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! How can I assist you today?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 15,
    "completion_tokens": 9,
    "total_tokens": 24
  }
}

In [9]:

# we set langfuse as a callback in the prev cell
# cohere call
response = completion(
  model="command-nightly",
  messages=[
    {"role": "user", "content": "Hi 👋 - i'm cohere"}
  ]
)

print(response)

{
  "object": "chat.completion",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "content": " Nice to meet you, Cohere! I'm excited to be meeting new members of the AI community",
        "role": "assistant",
        "logprobs": null
      }
    }
  ],
  "id": "chatcmpl-a14e903f-4608-4ceb-b996-8ebdf21360ca",
  "created": 1696392247.3313863,
  "model": "command-nightly",
  "usage": {
    "prompt_tokens": 8,
    "completion_tokens": 20,
    "total_tokens": 28
  }
}