The NVIDIA Run:ai Management Interface API is a centralized API which enables platform owners to programmatically manage tenants, clusters, and access across a multi-tenant environment. This API is designed for use by host organizations operating the centralized control plane and is not accessible to individual tenants.

NVIDIA Run:ai APIs are accessed using bearer tokens. To obtain a token, you need to create an Application using the api/v1/apps endpoint, which returns a clientId and clientSecret. Then, generate an access token using the /api/v1/token by sending a client_credentials grant request with those credentials.

To get your access token, follow the instructions in API Access Setup.