> ## 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.

# Call Anthropic on Bedrock

> Call Anthropic models via AWS Bedrock with Freeplay integration.

### 1. Configure Bedrock in Freeplay App

Configure Bedrock as a provider in your Freeplay account by going to
Models -> Toggle on Bedrock ->
Deploy a prompt with Bedrock

### 2. Import Anthropic Bedrock Client

Anthropic offers a specific Bedrock client in their SDK

### 3. Create a Bedrock Client

Instantiate your Bedrock client using the anthropic SDK

### 4. Fetch and Format Prompt

Retrieve a formatted prompt from freeplay

### 5. Call Bedrock Model

Call your Bedrock model

### 6. Record to Freeplay

Record the interaction back to freeplay

## Examples

<CodeGroup>
  ```python Python theme={null}
  from freeplay import Freeplay, RecordPayload, CallInfo
  from anthropic.lib.bedrock import AnthropicBedrock
  from anthropic import NotGiven
  import os
  import time
  from dotenv import load_dotenv

  load_dotenv()
  project_id = os.getenv("FREEPLAY_PROJECT_ID")
  freeplay_key = os.getenv("FREEPLAY_API_KEY")
  freeplay_url = os.getenv("FREEPLAY_URL")
  aws_access_key_id = os.getenv("AWS_ACCESS_KEY_ID")
  aws_secret_key = os.getenv("AWS_SECRET_KEY")

  fpClient = Freeplay(
      api_base=freeplay_url,
      freeplay_api_key=freeplay_key
  )

  bedrockClient = AnthropicBedrock(
      aws_secret_key=aws_secret_key,
      aws_access_key=aws_access_key_id,
      aws_region="us-east-1"
  )

  # fetch the prompt
  prompt_vars = {
      "pop_star": "Taylor Swift",
  }
  formatted_prompt = fpClient.prompts.get_formatted(
      project_id=project_id,
      template_name="album_bot",
      environment="latest",
      variables=prompt_vars
  )

  s = time.time()
  response = bedrockClient.messages.create(
      model=formatted_prompt.prompt_info.model,
      system=formatted_prompt.system_content or NotGiven(),
      messages=formatted_prompt.llm_prompt,
      **formatted_prompt.prompt_info.model_parameters
  )
  e = time.time()

  response_text = response.content[0].text

  all_messages = formatted_prompt.all_messages(
      {"role": "assistant",
      "content": response_text}
  )

  session = fpClient.sessions.create()

  payload = RecordPayload(
      project_id=project_id,
      all_messages=all_messages,
      inputs=prompt_vars,
      session_info=session,
      prompt_version_info=formatted_prompt.prompt_info,
      call_info=CallInfo.from_prompt_info(formatted_prompt.prompt_info, start_time=s, end_time=e)
  )
  fpClient.recordings.create(payload)
  ```
</CodeGroup>