This project provides an open-source, team-shared service for accessing various AI model APIs, primarily focusing on ChatGPT. It aims to enable safe sharing of API usage among team members, supporting multiple providers like OpenAI, Azure, Claude, and Gemini.

How It Works

The service acts as a centralized gateway for AI model requests. It allows users to share API keys and manage access, routing requests to the appropriate backend provider based on configuration. This approach simplifies API management for teams and enables cost-effective sharing of premium AI services.

Quick Start & Requirements

• Install/Run: Docker is the primary deployment method.
  • docker run -d --name opencatd -p 80:80 -v /etc/opencatd:/app/db mirrors2/opencatd-open
  • Docker Compose example provided.
• Prerequisites: Docker, optionally Nginx/Caddy/Traefik for HTTPS.
• Resources: Minimal resource footprint for basic Docker deployment.
• Links: Docker Compose

Highlighted Details

• Supports OpenAI (including Azure, Github Marketplace), Claude (VertexAI), and Gemini.
• Includes OpenAI's Whisper, TTS, and DALL-E capabilities.
• Provides commands for managing root tokens, saving/loading user data, and resetting tokens via docker exec.
• Offers guidance on custom host addresses and main page redirection via environment variables.

Maintenance & Community

• The project is undergoing a name change to "OpenTeam" and directs users to a new repository: mirrors2/openteam.
• No specific community links (Discord/Slack) or roadmap details are provided in the README.

Licensing & Compatibility

• The README does not explicitly state a license. The project name "openteam" and the description "开源实现" (open-source implementation) suggest an open-source nature, but the specific license terms are absent.

Limitations & Caveats

• The project is undergoing a name change, indicating potential ongoing development and possible breaking changes.
• The absence of a specified license makes commercial use and closed-source integration uncertain.
• HTTPS setup requires manual configuration with reverse proxies like Nginx or Caddy.