MULTI-PROVIDER LLM GATEWAY
One OpenAI-compatible API in front of every provider — with automatic failover, priority routing, streaming, and per-agent keys.
Works With
Live health and usage across all providers behind the gateway.
| Provider | Health | Last latency | Requests | Fail | Last error |
|---|
Upstream LLM backends, tried in priority order with automatic failover.
| Prio | Name | Type | Base URL | Key | Enabled | Actions |
|---|
Issue one key per agent. Your agents authenticate with these; the gateway uses its own provider keys behind the scenes.
Point any OpenAI-compatible client at the gateway and send your agent key.
| Name | Key | Rate limit | Requests | Last used | Enabled | Actions |
|---|
Send a test chat through the router and see which provider answered.
Most recent requests routed through the gateway.
| Time | Agent | Provider | Model | Strategy | Status | Latency | Tokens | Attempts |
|---|