Drop-in model routing

Change the base URL.
Keep your application.

RelayRouter exposes a familiar OpenAI-compatible API while routing requests across supported model providers and configured failover paths.

Get an API key Open documentation

A familiar request surface

curl https://relayrouter.org/v1/chat/completions \
  -H "Authorization: Bearer rr_live_••••" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-fable-5",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'

Streaming

Use server-sent events for responsive chat and generation interfaces.

Failover

Route models across multiple configured hosts without changing client code.

Metering

Track model, token, request, and cost data from the same operational workspace.

One endpoint. More leverage.

Compare models, create a scoped key, and move your first request.