leptonai
A Pythonic framework to simplify AI service building
What is leptonai?
A Pythonic framework to simplify AI service building
Overview of leptonai
What it is
Lepton AI is a Python library and command-line tool (lep) for operating the NVIDIA DGX Cloud Lepton platform. It provides a Pythonic framework to simplify AI service building and management through code and CLI commands.
Key features
- A
lepCLI tool to create and manage endpoints, Dynamo graph deployments, batch jobs, dev pods, Ray clusters, fine-tuning jobs, storage, secrets, and more, plus inspect managed Slurm clusters and jobs. - Pythonic configuration specs that can be shipped directly to the cloud.
- Skills that allow agents to operate the Lepton platform.
- Support for Teleport SSH into pods, jobs, and Slurm compute nodes (requires
tshv18 or newer). - Authentication controls including API tokens, unauthenticated access opt-out, and token redaction in output.
- Interactive shell sessions tunneled through the workspace API over HTTPS, requiring no SSH key, public IP, or Teleport client.
Use cases
- Deploying container images as endpoints with authentication and IP access controls.
- Running batch jobs and launching interactive dev pods for development and training.
- Inspecting and managing managed Slurm clusters and jobs.
- Connecting to pods, endpoints, jobs, and Ray clusters via shell or Teleport SSH for debugging and administration.
Getting started
Install the library and the lep CLI: shell pip install -U leptonai To install the latest development version from the main branch: shell pip install -U 'git+https://github.com/leptonai/leptonai.git@main' Log in to your workspace: shell lep login Deploy a container image as an endpoint: shell lep endpoint create -n my-endpoint --container-image my-registry/my-app:latest lep endpoint list lep endpoint status -n my-endpoint
License and pricing
The project is licensed under Apache-2.0 and is open source.
Based on the project's public README and repository metadata.