Summary This project offers an automated service that bridges Google Gemini Business with OpenAI-compatible APIs. It targets developers and power users needing to integrate Gemini's capabilities into existing workflows, providing automated account management, token refreshing, and a unified API endpoint. The core benefit is simplified access to Gemini's advanced features, including multi-modal processing and image generation, without manual intervention.

How It Works The system acts as a proxy, translating requests to an OpenAI-compatible format and routing them to managed Gemini Business accounts. It utilizes Docker with Chrome/Xvfb for automated browser tasks like registration and login refreshes. Key features include robust multi-account management with load balancing, automatic failover, intelligent circuit breaking for failed accounts, and a three-layer retry strategy for high availability. A web-based admin panel offers real-time monitoring and configuration.

Quick Start & Requirements Deployment options include Docker Compose (Linux/amd64 only), Docker build, or local Python/uv.

• Docker Compose: Download docker-compose.yml, configure environment variables, and run docker-compose up -d.
• Docker Build: Clone repo, build image (docker build), and run container (docker run).
• Local: Install dependencies (uv sync), configure .env, and run uv run main.py. Prerequisites: Docker, Python 3.11+, and specific environment variables for auto-registration (temporary email API, Google login URL).

Highlighted Details

• Multi-Account Management: Load balancing, failover, circuit breaking (429/error disabling), triple retries, auto-disable on expiration, manual disable/enable.
• Automated Operations: One-click batch registration, auto email verification, cookie/token refresh, browser process crash recovery.
• Extensive Multi-modal Support: Handles 100+ file types (images, PDFs, Office, audio, video, code).
• Image Capabilities: Supports gemini-3-pro-preview for image generation and image-to-image tasks.
• Admin Panel: Real-time monitoring, usage stats, online configuration management.

Maintenance & Community No explicit maintainers or direct community links (Discord/Slack) are listed. Discussions are referenced via external platforms like Linux.do.

Licensing & Compatibility Released under the MIT License, which is generally permissive for commercial use and integration into closed-source projects.

Limitations & Caveats Docker Compose deployment is restricted to linux/amd64 architectures. Auto-registration setup requires external email services and Google login URLs. Generated images and in-memory logs are temporary and may be lost on restart unless specific configurations are used. Configuration persistence across restarts may require syncing environment variables with saved JSON files.