Lightning SDK

The Lightning SDK lets you programmatically interact with Lightning AI Studios: run jobs, manage teamspaces, attach machines, deploy models, and more — all from Python or the CLI.

Start Here

Use this path if you are new to the Lightning SDK:

1. Installation
2. Quickstart

Key Concepts

• Manage Studio instances — create, start, stop, and run commands
• Submit and track jobs with lightning_sdk.Job.run and Machine selection
• Organize work inside a Teamspace
• Scale training with lightning_sdk.MMT.run
• Deploy models with Deployment

Quick Example

from lightning_sdk import Studio, Machine

studio = Studio("my-studio")
studio.start()

job = studio.run("python train.py", machine=Machine.A10G)
job.wait()

studio.stop()

Documentation Map

Install — set up the package and authenticate.

Quickstart — run your first job in minutes.

The Examples provide tutorial walkthroughs for common CLI and SDK workflows.

The API reference documents every public class and function.

Start

• Installation
• Quickstart
• Examples
• API Reference