Skip to content


Developers can access Run:ai through various programmatic interfaces.

API Support

The endpoints and parameters specified in the API reference are the ones that are officially supported by Run:ai. Endpoints and parameters that are NOT listed in the reference are not officially supported.

API endpoints and fields marked as deprecated remain operational and can be used; however, Run:ai will NO LONGER RECOMMEND USING THEM, WILL NOT ADD FUNCTIONALITY TO THEM, and WILL STOP SUPPORTING THEM after 2 major releases (for self-hosted deployments), and after 6 months of support for SaaS deployments.

For details, see the Deprecation notifications.

API Architecture

Run:ai is composed of a single, multi-tenant control plane. Each tenant can be connected to one or more GPU clusters. See Run:ai system components for detailed information.

Below is a diagram of the Run:ai API Architecture. A developer may:

  1. Access the control plane via the Administrator API.
  2. Access any one of the GPU clusters via Cluster API.
  3. Access cluster metrics via the Metrics API.

api architecture image

Administrator API

Add, delete, modify and list Run:ai meta-data objects such as Projects, Departments, Users, and more.

The API is provided as REST and is accessible via the control plane endpoint.

For more information see Administrator REST API.

Cluster API

Submit and delete Workloads.

The API is provided as Kubernetes API.

Cluster API is accessible via the GPU cluster itself. As such, multiple clusters may have multiple endpoints.


The same functionality is also available via the Run:ai Command-line interface. The CLI provides an alternative for automating with shell scripts.

Metrics API

Retrieve metrics from multiple GPU clusters.

See the Metrics API document.

API Authentication

See API Authentication for information on how to gain authenticated access to Run:ai APIs.