Model Pools
Aggregate models into virtual pools. Round-robin, failover, rate limiting.
Streaming & Live
SSE streaming with TTFT. Real-time active request monitoring.
Metrics & Stats
Latency, tokens, tokens/sec. Per-provider and per-model breakdown.
API Key Auth
Secure key-based authentication. HMAC-signed sessions for web panel.
Provider Proxy
HTTPS/SOCKS5 proxy per provider. IP masking, bypass blocks.
Rate Limiting
Minimum interval between provider requests. In-memory limiter, resets on restart.
OpenAI-compatible API
Drop-in replacement. Change the provider — the client stays the same.
curl https://your-server/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer sk-..." \
-d '{
"model": "vitold/free",
"messages": [
{"role": "user", "content": "Hello!"}
]
}'How it works
1. Add providers (OpenRouter, Together, Groq...)
2. Create a model pool with your preferred strategy
3. Generate an API key and use it in your client