🔄 Gemini-API-Rotator

A high-performance proxy server that automatically rotates between multiple Google Gemini API keys to avoid rate limits. Perfect for applications requiring high-throughput access to Gemini API.

✨ Features

• 🔗 Customizable API Base URL: Easily switch between different Gemini API endpoints
• 🔑 Automatic Key Rotation: Seamlessly switches between multiple API keys
• 🚀 High Performance: Handles both streaming and regular requests efficiently
• 🛡️ Smart Rate Limit Handling: Automatically detects and bypasses rate-limited keys
• 📊 Intelligent Load Balancing: Remembers last successful key for optimal performance
• 🔌 Drop-in Replacement: Works with any Gemini API client
• 📝 Comprehensive Logging: Detailed logs for monitoring and debugging

🚀 Quick Start

Installation

# Clone the repository
git clone https://github.com/smitkunpara/gemini-proxy-rotator.git
cd gemini-proxy-rotator

# Install dependencies
pip install -r requirements.txt

# Copy environment example
cp .env.example .env

Configuration

Edit .env file and add your Gemini API keys:

# Using JSON array format (recommended)
GEMINI_API_KEYS='["your-api-key-1", "your-api-key-2", "your-api-key-3"]'

# OR using comma-separated format
GEMINI_API_KEYS=your-api-key-1,your-api-key-2,your-api-key-3

Running the Server

# Using Python directly
python -m gemini-proxy

# With custom port
python -m gemini-proxy --port 8888 --log-level DEBUG

📖 Usage

Once the proxy is running, simply point your Gemini API client to http://localhost:8080 instead of the original Gemini API base URL. You can also configure a custom Gemini API base URL using the GEMINI_API_BASE_URL environment variable.

⚙️ Advanced Configuration

Environment Variable	Description	Default
GEMINI_API_KEYS	JSON array or comma-separated list of API keys	Required
GEMINI_API_BASE_URL	Base URL for the Gemini API (e.g., https://generativelanguage.googleapis.com)	https://generativelanguage.googleapis.com
PROXY_PORT	Port to run the proxy server	8080
PROXY_HOST	Host to bind the server	0.0.0.0
LOG_LEVEL	Logging level (DEBUG, INFO, WARNING, ERROR)	INFO
CHUNK_SIZE	Streaming chunk size in bytes	1024
STREAM_TIMEOUT	Timeout for streaming requests (seconds)	300
REQUEST_TIMEOUT	Timeout for regular requests (seconds)	60

🔧 How It Works

1. Request Interception: The proxy intercepts all requests meant for Gemini API
2. Key Selection: Starts with the last successful key or rotates if needed
3. Request Forwarding: Forwards the request with the selected API key
4. Error Handling: If a key is rate-limited or fails, automatically tries the next key
5. Response Streaming: Efficiently streams responses for streaming endpoints
6. Success Tracking: Remembers successful keys for optimal performance

🤝 Contributing

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

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Built to solve rate limiting issues with Google's Gemini API

⚠️ Disclaimer

This tool is meant for legitimate use cases where you have multiple API keys for load balancing. Please ensure you comply with Google's Gemini API terms of service.