Router Documentation

How the pieces fit

Your application talks only to Router. Router checks each request, selects eligible capacity, and relays the provider's output back to you.

Your model, your code

Use an OpenAI client or a simple HTTP request. Choose a supported model, send a message, and receive a complete answer or stream it into your application.

1ROUTER_REQUEST_ID="$(uuidgen)" # Reuse this ID if you retry the same request.2curl "$ROUTER_BASE_URL/chat/completions" \3  -H "Authorization: Bearer $ROUTER_API_KEY" \4  -H "Idempotency-Key: $ROUTER_REQUEST_ID" \5  -H "X-Max-Charge-Credits: 5.000000" \6  -H "Content-Type: application/json" \7  -d '{8    "model": "gemma3:1b",9    "messages": [10      { "role": "user", "content": "Hello!" }11    ]12  }'

Set ROUTER_BASE_URL to the endpoint shown in Developer API, including /v1. Keep ROUTER_API_KEY in your server environment.

A few useful starting points

Build a reliable integration

• Choose a model and validate its current edition limits.
• Use Auto or choose a provider at a reviewed pricing revision.
• Design messages that stay within the supported text surface.
• Stream responses while preserving clear completion and error states.
• Add bounded retries and cancellation.
• Correlate failures with usage and observability.
• Parse the complete error response contract.