Skip to main content
You aren’t limited to Opper’s hosted models. Register your own provider deployment, for example a private Azure OpenAI deployment, with its API key. It then appears in the catalog alongside built-in models and gets the same routing, governance, and tracing.

Register a custom model

The quickest way is in the dashboard. Open Models in platform.opper.ai, choose Create → BYOK model, and fill in: Save it, then call the model by its name like any other:
Custom models belong to your organization, and each name can be used once in it. Pick a name no catalog model uses: when a name matches both, the catalog model is the one called.
A custom model can’t be called while a model access rule or a lock applies to the call, because its location and data handling are only what you declare. Registering still succeeds, and the call then answers 403.

What your endpoint needs

An endpoint you set must use HTTPS, and its hostname must resolve only to public IP addresses. http:// URLs, localhost, and private or link-local addresses are refused, so a server on a private network has to be reachable at a public HTTPS address first. Registering makes one short test call to the model with your credentials. If that call fails, you get a 400 with the reason and nothing is saved.

Register from the API or CLI

If you’d rather automate it, the API takes these fields: The CLI takes name, identifier, and api_key as arguments, and extra as --extra.

What’s next

Models

The full catalog of built-in models.

AI Gateway

How routing and model selection fit together.
Last modified on November 9, 2026