AI Gateway

Configure a model route with a selected provider, model, fallback, and request controls, then invoke its chat-completions endpoint.

Open AI Gateway

Configure a route

  1. Create a route with a valid slug and a provider/model available to the deployment.
  2. Review fallback, timeout, request/token limits, and input-size settings.
  3. Save the route and copy its generated invocation path.
  4. Send a small non-streaming chat request and inspect the returned status, completion, and diagnostic headers.

Access and limits

The route uses the provider credentials configured on the server. Account API keys and provider API keys serve different purposes; use the account credential accepted by the invocation endpoint.

The route pins its approved primary model and can attempt the configured fallback after a provider failure. Limits are the configured route values, not a universal Premium or Enterprise quota. Streaming is rejected by the current chat-completions handler.

Troubleshooting

Check account and service access for 401/403, the route slug for 404, input size for 413, configured budgets for 429, and provider/runtime configuration for 5xx responses. Keep the request ID when asking for support.

API endpoints

Use your Genorbex deployment’s origin followed by the path below. Replace IDs with resources accessible to your account. Authentication and permissions are checked by each endpoint.

POST /api/orbexdb/ai-ml/gateway/{slug}/v1/chat/completions

Invokes the configured route’s provider/model policy.

Authentication: Genorbex account token and service/resource access; API-key support follows the endpoint’s authentication guard.

Example JSON body

{
  "messages": [
    {
      "role": "user",
      "content": "Summarize this short note."
    }
  ],
  "max_tokens": 200
}

Need help with your setup? Visit the Help Center or contact Genorbex.