Skip to main content

AI Runtime CLI

Beta

The AI Runtime CLI is in Beta.

The air command-line interface submits and manages distributed training workloads on AI Runtime, the on-demand serverless GPU compute platform. The CLI uses YAML-based job configuration, integrates with MLflow, and supports workspace-based and git-based code workflows.

When to use the CLI

Use the AI Runtime CLI when you want to:

  • Submit GPU training workloads from your laptop and code editor without opening a notebook.
  • Define training jobs declaratively in YAML so they can be checked into source control.

For the in-notebook Python API (@distributed and @ray_launch), see Multi-GPU workload instead.

In this section