Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.routing.run/llms.txt

Use this file to discover all available pages before exploring further.

System status

GET /v1/status. No authentication required.
curl -sS https://api.routing.run/v1/status
Response shape (values change every request):
{
  "providers": [
    { "name": "upstream-a", "status": "degraded", "latency_ms": 904 },
    { "name": "upstream-b", "status": "healthy", "latency_ms": 153 }
  ],
  "models": [
    {
      "id": "route/deepseek-v3.2",
      "name": "DeepSeek V3.2",
      "tier": "free",
      "status": "online"
    }
  ],
  "incidents": [
    {
      "id": "inc-001",
      "title": "OpenRouter Rate Limiting",
      "description": "Some OpenRouter models are experiencing rate limiting due to high demand.",
      "severity": "warning",
      "status": "ongoing",
      "created_at": "2026-04-15T08:00:00+00:00",
      "resolved_at": null
    },
    {
      "id": "inc-002",
      "title": "Trinity Large Preview Unavailable",
      "description": "The arcee-ai/trinity-large-preview model endpoint is not responding.",
      "severity": "error",
      "status": "resolved",
      "created_at": "2026-04-15T05:00:00+00:00",
      "resolved_at": "2026-04-15T09:00:00+00:00"
    }
  ],
  "last_updated": "2026-04-15T10:19:52.931465+00:00"
}
Use this endpoint as a public health check in apps, agents, and monitoring. It is useful for deciding whether routing.run is reachable before you debug a request flow. The top-level providers array is synthetic health from routing.run probes. models is built from published routing config and display metadata. incidents in the current server build is hard-coded sample data (always two items: inc-001 ongoing warning, inc-002 resolved error) baked into status.py — do not treat it as live incident feed. Prefer status.routing.run for real operator status until the API reads external incident data. last_updated is datetime.now(UTC).isoformat() from the API host.

Health states

StatusMeaning
healthyService is responding normally
degradedService is responding with increased latency or errors
unknownNo recent health data is available

Public settings

GET /v1/settings — public configuration:
curl -sS https://api.routing.run/v1/settings
Response:
{
  "cost_multiplier": 1.65
}

Status page

For real-time status updates and incident history, visit status.routing.run.