Integration

RunPod Serverlesswith Meridian as routing layer

Deploy GPU workloads on RunPod's serverless infrastructure and let Meridian handle authentication, rate limiting, and request routing — so your inference endpoints stay fast, secure, and observable without touching your worker code.

Auth gateway

API keys validated at the edge before requests reach your workers.

Smart routing

Load balance across RunPod regions with automatic failover.

Usage analytics

Per-endpoint metrics, cold-start tracking, and cost attribution.

Quick start

  1. 1Deploy your worker on RunPod Serverless and note the endpoint URL.
  2. 2Register the upstream in Meridian under Integrations → RunPod.
  3. 3Point your clients at the Meridian proxy URL — auth, routing, and telemetry are live.