Liara_Chat_Completion_Proxy

🚀 Universal AI Proxy

Professional AI Model Proxy Gateway with support for 100+ models, multi-provider load balancing, secure user management, and detailed usage tracking.

✨ Key Features

• 🤖 100+ Models Support: Powered by LiteLLM, supporting OpenAI, Anthropic, Google, Hugging Face, and more.
• 🛡️ Secure Auth System: JWT-based authentication with SHA-256 hashed API keys.
• 📊 Advanced Analytics: Database-side token usage and cost calculation (PostgreSQL/SQLite).
• 🌐 Proxy & Load Balancer: HTTP/SOCKS5 proxy support with country-based routing and priority failover.
• 🕹️ Professional Dashboard: Modern dark-themed UI (Vue.js) with 100% locally bundled assets for offline use.
• ⚡ Non-Blocking Performance: Fully asynchronous architecture using litellm.acompletion.
• 🔄 DevOps Ready: Automated tagging, GitHub Releases, and GitLab mirroring.

🚀 Quick Start

1. Install Dependencies:

   pip install -r requirements.txt
   

2. Environment Setup: Create a .env file:

   DATABASE_URL=sqlite:///./sql_app.db
   SECRET_KEY=your_secure_random_key
   

3. Run the Server:

   uvicorn main:app --host 0.0.0.0 --port 8100
   

4. Access the Dashboard: Open http://localhost:8100 in your browser. (The first registered user becomes Admin).

🛠 Admin Capabilities

• Providers: Add multiple API keys per provider with priority levels.
• Proxies: Manage HTTP/SOCKS5 proxy lists for IP protection.
• Usage: Monitor system-wide usage, costs, and token consumption.

🤖 API Usage (OpenAI Compatible)

curl http://localhost:8100/api/v1/chat/completions \
  -H "Authorization: Bearer YOUR_SECURE_KEY" \
  -d '{
    "model": "openai/gpt-4o",
    "messages": [{"role": "user", "content": "Hello world"}]
  }'

📜 Development & Contributions

Please see CONTRIBUTING.md and CHANGELOG.md for details.

---

Developed by MovtiGroup

This site is open source. Improve this page.