Use Anthropic clients (like Claude Code) with Gemini or OpenAI backends. 🤝

A proxy server that lets you use Anthropic clients with Gemini or OpenAI models via LiteLLM. 🌉

• OpenAI API key 🔑
• Google AI Studio (Gemini) API key (if using default provider) 🔑
• uv installed.

1. Clone this repository:

   git clone https://github.com/1rgs/claude-code-openai.git
   cd claude-code-openai

2. Install uv (if you haven't already):

   curl -LsSf https://astral.sh/uv/install.sh | sh

   (uv will handle dependencies based on pyproject.toml when you run the server)

3. Configure Environment Variables: Copy the example environment file:

   cp .env.example .env

   Edit .env and fill in your API keys and model configurations:

   • ANTHROPIC_API_KEY: (Optional) Needed only if proxying to Anthropic models.
   • OPENAI_API_KEY: Your OpenAI API key (Required if using OpenAI models as fallback or primary).
   • GEMINI_API_KEY: Your Google AI Studio (Gemini) API key (Required if using the default Gemini preference).
   • PREFERRED_PROVIDER (Optional): Set to google (default) or openai. This determines the primary backend for mapping haiku/sonnet.
   • BIG_MODEL (Optional): The model to map sonnet requests to. Defaults to gemini-2.5-pro-preview-03-25 (if PREFERRED_PROVIDER=google and model is known) or gpt-4o.
   • SMALL_MODEL (Optional): The model to map haiku requests to. Defaults to gemini-2.0-flash (if PREFERRED_PROVIDER=google and model is known) or gpt-4o-mini.

   Mapping Logic:

   • If PREFERRED_PROVIDER=google (default), haiku/sonnet map to SMALL_MODEL/BIG_MODEL prefixed with gemini/ if those models are in the server's known GEMINI_MODELS list.
   • Otherwise (if PREFERRED_PROVIDER=openai or the specified Google model isn't known), they map to SMALL_MODEL/BIG_MODEL prefixed with openai/.

4. Run the server:

   uv run uvicorn server:app --host 0.0.0.0 --port 8082 --reload

   (--reload is optional, for development)

1. Install Claude Code (if you haven't already):

   npm install -g @anthropic-ai/claude-code

2. Connect to your proxy:

   ANTHROPIC_BASE_URL=http://localhost:8082 claude

3. That's it! Your Claude Code client will now use the configured backend models (defaulting to Gemini) through the proxy. 🎯

The proxy automatically maps Claude models to either OpenAI or Gemini models based on the configured model:

The following OpenAI models are supported with automatic openai/ prefix handling:

• o3-mini
• o1
• o1-mini
• o1-pro
• gpt-4.5-preview
• gpt-4o
• gpt-4o-audio-preview
• chatgpt-4o-latest
• gpt-4o-mini
• gpt-4o-mini-audio-preview

The following Gemini models are supported with automatic gemini/ prefix handling:

• gemini-2.5-pro-preview-03-25
• gemini-2.0-flash

The proxy automatically adds the appropriate prefix to model names:

• OpenAI models get the openai/ prefix
• Gemini models get the gemini/ prefix
• The BIG_MODEL and SMALL_MODEL will get the appropriate prefix based on whether they're in the OpenAI or Gemini model lists

For example:

• gpt-4o becomes openai/gpt-4o
• gemini-2.5-pro-preview-03-25 becomes gemini/gemini-2.5-pro-preview-03-25
• When BIG_MODEL is set to a Gemini model, Claude Sonnet will map to gemini/[model-name]

You can customize which models are used via environment variables:

• BIG_MODEL: The model to use for Claude Sonnet models (default: "gpt-4o")
• SMALL_MODEL: The model to use for Claude Haiku models (default: "gpt-4o-mini")
• PREFERRED_PROVIDER: Set to "google" (default), "openai" or "custom" to choose the primary backend

Add these to your .env file to customize:

OPENAI_API_KEY=your-openai-key
# For OpenAI models
PREFERRED_PROVIDER=openai
BIG_MODEL=gpt-4o
SMALL_MODEL=gpt-4o-mini

# For Gemini models
PREFERRED_PROVIDER=google
BIG_MODEL=gemini-2.5-pro-preview-03-25
SMALL_MODEL=gemini-2.0-flash

# For custom OpenAI-compatible models
PREFERRED_PROVIDER=custom
BIG_MODEL=my-large-model
SMALL_MODEL=my-small-model

You can add support for custom OpenAI-compatible models by creating a custom_models.yaml file:

- model_id: "my-model"
  api_base: "https://api.example.com"
  api_key_name: "MY_API_KEY"
  can_stream: true
  max_tokens: 8192
  model_name: "actual-model-name"  # Optional - if different from model_id

Key features of custom models:

• Supports any OpenAI-compatible API endpoint
• Handles streaming responses if can_stream: true
• Automatically adds custom/ prefix to model names
• Uses specified API key from environment variables
• Supports tool use (function calling) if the backend supports it
• Maintains Anthropic API compatibility while using custom backends
• Automatically loads from custom_models.yaml on server startup
• Supports multiple custom models in the same configuration file

Then set these environment variables:

PREFERRED_PROVIDER=custom
BIG_MODEL=my-model
MY_API_KEY=your-api-key

Or set them directly when running the server:

# Using OpenAI models (with uv)
BIG_MODEL=gpt-4o SMALL_MODEL=gpt-4o-mini uv run uvicorn server:app --host 0.0.0.0 --port 8082

# Using custom models (with uv)
PREFERRED_PROVIDER=custom BIG_MODEL=my-large-model SMALL_MODEL=my-small-model uv run uvicorn server:app --host 0.0.0.0 --port 8082

# Using Gemini models (with uv)
BIG_MODEL=gemini-2.5-pro-preview-03-25 SMALL_MODEL=gemini-2.0-flash uv run uvicorn server:app --host 0.0.0.0 --port 8082

# Using custom models (with uv)
PREFERRED_PROVIDER=custom BIG_MODEL=my-large-model SMALL_MODEL=my-small-model uv run uvicorn server:app --host 0.0.0.0 --port 8082

This proxy works by:

1. Receiving requests in Anthropic's API format 📥
2. Translating the requests to OpenAI format via LiteLLM 🔄
3. Sending the translated request to OpenAI 📤
4. Converting the response back to Anthropic format 🔄
5. Returning the formatted response to the client ✅

The proxy handles both streaming and non-streaming responses, maintaining compatibility with all Claude clients. 🌊

Contributions are welcome! Please feel free to submit a Pull Request. 🎁