8Router

AI Routing Gateway — One Router. Multi Providers. Zero Downtime.

Features

• 🚀 20+ Providers — Groq, Mistral, OpenRouter, Ollama, and more
• 🔄 3-Tier Fallback — Automatic failover between providers
• 💰 Token Saver — 20-65% cost reduction via smart routing
• 🎯 9 Service Kinds — Coding, Creative, Privacy, Fast, Smart, etc.
• 📊 Dashboard — Real-time monitoring and analytics
• 🔑 API Keys — Virtual key management with quotas
• 🛡️ Guardrails — Content filtering and rate limiting

Quick Start

# Install globally
npm install -g 8router

# Start the server
8router

# Or run with custom config
8router --config ./my-config.yaml

API Endpoints

• API: http://localhost:8080/v1
• Dashboard: http://localhost:8081
• Hermes Proxy: http://localhost:8083/v1

Configuration

Create 8router.yaml in your project root:

providers:
  - id: groq
    apiKey: your-groq-key
    models: ["llama-3.1-70b-versatile"]
  
  - id: mistral
    apiKey: your-mistral-key
    models: ["mistral-large-latest"]

routing:
  strategy: fallback
  timeout: 30000

CLI Tools

8Router works with:

• Hermes Agent — Nous Research AI agent
• Claude Code — Anthropic's CLI
• OpenAI Codex — OpenAI's coding CLI
• Cursor — AI-first code editor
• Cline — VS Code AI assistant

Dashboard

Access the dashboard at http://localhost:8081 to:

• Monitor request volume and latency
• Track token usage and costs
• Test providers in the Playground
• Manage API keys and quotas
• View system health and logs

License

MIT © 8Agents