Private AI on your terms.

Run open-weight models through an OpenAI-compatible API, manage keys and projects in AI Studio, and deploy on a private cloud you control.

New to Nebul? Sign up

Browse the docs

Get started

From an API key to a running model.

Make your first request

Copy-paste examples that run against the Nebul Inference API.
  1. 1. Get an API key

    Create a Nebul account and issue a key in AI Studio.

  2. 2. Send your first request

    Run a cURL or Python example against the API.

  3. 3. Choose the right model

    Compare models by capability, context window, and price.

  4. 4. Prepare for production

    Rate limits, scaling, and request tracing.

Learn & reference

Endpoints, integrations, and release notes.