> ## Documentation Index
> Fetch the complete documentation index at: https://docs.freeplay.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Provider Switching with LiteLLM

> Route LLM calls through LiteLLM while recording to Freeplay.

### 1. Create the Freeplay Client

### 2. Fetch the prompt from Freeplay

Freeplay will handle conversion of the messages to the right format for you

### 3. Call the LLM via LiteLLM

Route the LLM call through LiteLLM

### 4. Record to Freeplay

Record the interaction back to Freeplay

## Examples

<CodeGroup>
  ```python Python theme={null}
  from freeplay import Freeplay, RecordPayload, CallInfo, SessionInfo
  from openai import OpenAI
  from anthropic import Anthropic
  from dotenv import load_dotenv
  import os
  import time
  from litellm import completion

  load_dotenv("../.env")

  fp_client = Freeplay(
      freeplay_api_key=os.getenv("FREEPLAY_API_KEY"),
      api_base=os.getenv("FREEPLAY_URL"),
  )

  user_question = "What is the capital of France?"
  # Fetch the prompt from freeplay
  prompt_vars = {"question": user_question}
  formatted_prompt = fp_client.prompts.get_formatted(
      project_id="5688ebaf-7f22-4d5d-b9bb-bc715c8faabb",
      template_name="BasicTriviaBot",
      environment="dev",
      variables=prompt_vars,
  )

  start = time.time()

  response = completion(
    model=f"{formatted_prompt.prompt_info.provider}/{formatted_prompt.prompt_info.model}",
    messages=formatted_prompt.llm_prompt,
  )
  msg = response.choices[0].message
  answer = msg.content

  end = time.time()

  # record to Freeplay
  session = fp_client.sessions.create()
  fp_client.recordings.create(
      RecordPayload(
  				project_id=project_id,
          all_messages=formatted_prompt.all_messages(msg),
          inputs=prompt_vars,
          session_info=session,
          prompt_version_info=formatted_prompt.prompt_info,
          call_info=CallInfo.from_prompt_info(formatted_prompt.prompt_info, start, end)
      )
  )

  print(f'Question {user_question} \n Answer {answer}')
  ```
</CodeGroup>