Skip to content

Viewing and Submitting Deployments

The Run:ai User interface Deployment area allows the viewing and submitting of Deployments for serving inference workloads. Submitting inference workloads can only be done if your user has ML Engineer access.

Deployment list

The main view shows a list of Deployments:

job-list

Submit a Deployment

On the top right, you can choose to Submit a new Deployment.

Note

If knative is not installed in your cluster the button will be grayed out.

A Deployment form will open:

submit-job

Note

If the Deploy button is disabled or does not exist, then your cluster is not installed or configured to connect to the cluster see here for more information.

Deployment Properties

When selecting a single Deployment, a right-pane appears:

job-properties

This multi-tab view provides information about Deployment details, related Pods, Deployment status history, and various utilization graphs.