Anuneko OpenAI Proxy

简体中文

A lightweight proxy that converts Anuneko services into a standard OpenAI API format.

✨ Features

• Full Compatibility: Supports both streaming and non-streaming OpenAI responses.
• Smart Context: Automatically manages upstream chat_id based on conversation history for seamless continuity.
• Multi-Model: Switch between "Orange Cat" and "Exotic Shorthair" via model names.
• Access Control: Secure your proxy with custom API Keys.

🚀 Deployment

1. Create a .env file:

   ANUNEKO_TOKEN=your_upstream_token
   PROXY_API_KEYS=["sk-your-api-key"]

2. Build and run the container:

   docker build -t anuneko-proxy .

   docker run -d \
     --name my-anuneko \
     -p 8000:8000 \
     --env-file .env \
     anuneko-proxy

3. OR run directly:

   python -m app.main
   

⚙️ Configuration

Environment Variables (.env)

Variable	Description	Example
ANUNEKO_TOKEN	Required. The x-token from Anuneko	eyJh...
ANUNEKO_COOKIE	Optional. Required by some accounts	session=...
PROXY_API_KEYS	List of valid keys for clients	["sk-123456"]

Client Setup (e.g., NextChat)

• Endpoint: http://your-server-ip:8000
• API Key: The value you set in PROXY_API_KEYS (e.g., sk-123456).
• Custom Models: orange-cat (Default), black-cat (Exotic Shorthair).

⚠️ Disclaimer

This project is for educational and research purposes only.

1. This is a third-party wrapper; no data is stored locally. All requests are forwarded upstream.
2. Do not use this project for commercial or illegal purposes.
3. The developers are not responsible for any account bans or damages resulting from the use of this tool.

📄 License

This project is licensed under the GNU General Public License v3.0 (GPLv3).