Skip to content

README.md

Latest commit

 

History

History
72 lines (45 loc) · 1.74 KB

README.md

File metadata and controls

72 lines (45 loc) · 1.74 KB

Gemini CLI Proxy Server

OpenAI-compatible API wrapper for Google's Gemini models.

License Status

This proxy allows you to use Google's Gemini models with any tool, library, or script designed for OpenAI's API. It handles authentication, stream adaptation, and multimodal input conversion automatically.

📚 Documentation

We have comprehensive documentation available in the docs/ directory:

🚀 Introduction

What is this project and why should I use it?

🔐 Authentication

How do I secure my proxy and connect to Google?

  • Consumer Auth: Protect your proxy with PROXY_API_KEY.
  • Provider Auth: Use API Keys, OAuth (Local), or Vertex AI.

🔌 API Reference

Endpoints, parameters, and cURL examples.

🐳 Deployment

How to host this server anywhere.

🏗️ Architecture

Deep dive into how the proxy works internally.

🔧 Troubleshooting

Common errors and solutions.

Quick Start (Local)

  1. Install & Build:

    npm install
npm run build

  2. Run:

    # With Google AI Studio Key
export GEMINI_API_KEY="AIzaSy..."
npm run start

  3. Test:

    curl http://localhost:3000/v1/models