Skip to main content

Claude Code

This tutorial shows how to call the Responses API models like codex-mini and o3-pro from the Claude Code endpoint on LiteLLM.

info

This tutorial is based on Anthropic's official LiteLLM configuration documentation. This integration allows you to use any LiteLLM supported model through Claude Code.

Prerequisites​

  • Claude Code installed
  • API keys for your chosen providers

Installation​

First, install LiteLLM with proxy support:

pip install 'litellm[proxy]'

1. Setup config.yaml​

Create a secure configuration using environment variables:

model_list:
  # Responses API models
  - model_name: codex-mini    
    litellm_params:
      model: openai/codex-mini
      api_key: os.environ/OPENAI_API_KEY
      api_base: https://api.openai.com/v1
  
  - model_name: o3-pro
    litellm_params:
      model: openai/o3-pro
      api_key: os.environ/OPENAI_API_KEY
      api_base: https://api.openai.com/v1

litellm_settings:
  master_key: os.environ/LITELLM_MASTER_KEY

Set your environment variables:

export OPENAI_API_KEY="your-openai-api-key"
export LITELLM_MASTER_KEY="sk-1234567890"  # Generate a secure key

2. Start proxy​

litellm --config /path/to/config.yaml

# RUNNING on http://0.0.0.0:4000

3. Verify Setup​

Test that your proxy is working correctly:

curl -X POST http://0.0.0.0:4000/v1/messages \
-H "Authorization: Bearer $LITELLM_MASTER_KEY" \
-H "Content-Type: application/json" \
-d '{
    "model": "codex-mini",
    "messages": [{"role": "user", "content": "What is the capital of France?"}]
}'

4. Configure Claude Code​

Setup Claude Code to use your LiteLLM proxy:

export ANTHROPIC_BASE_URL="http://0.0.0.0:4000"
export ANTHROPIC_AUTH_TOKEN="$LITELLM_MASTER_KEY"

5. Use Claude Code​

Start Claude Code with any configured model:

# Use Responses API models
claude --model codex-mini
claude --model o3-pro

# Or use the latest model alias
claude --model codex-mini-latest

Example conversation:

Troubleshooting​

Common issues and solutions:

Claude Code not connecting:

  • Verify your proxy is running: curl http://0.0.0.0:4000/health
  • Check that ANTHROPIC_BASE_URL is set correctly
  • Ensure your ANTHROPIC_AUTH_TOKEN matches your LiteLLM master key

Authentication errors:

  • Verify your environment variables are set: echo $LITELLM_MASTER_KEY
  • Check that your OpenAI API key is valid and has sufficient credits

Model not found:

  • Ensure the model name in Claude Code matches exactly with your config.yaml
  • Check LiteLLM logs for detailed error messages

Using Multiple Models​

Expand your configuration to support multiple providers and models:

model_list:
  # Responses API models
  - model_name: codex-mini
    litellm_params:
      model: openai/codex-mini
      api_key: os.environ/OPENAI_API_KEY
      api_base: https://api.openai.com/v1
  
  - model_name: o3-pro
    litellm_params:
      model: openai/o3-pro
      api_key: os.environ/OPENAI_API_KEY
      api_base: https://api.openai.com/v1

  # Standard models
  - model_name: gpt-4o
    litellm_params:
      model: openai/gpt-4o
      api_key: os.environ/OPENAI_API_KEY

  - model_name: claude-3-5-sonnet
    litellm_params:
      model: anthropic/claude-3-5-sonnet-20241022
      api_key: os.environ/ANTHROPIC_API_KEY

litellm_settings:
  master_key: os.environ/LITELLM_MASTER_KEY

Switch between models seamlessly:

# Use Responses API models for advanced reasoning
claude --model o3-pro
claude --model codex-mini

# Use standard models for general tasks
claude --model gpt-4o
claude --model claude-3-5-sonnet