Welcome to Lepton

Lepton AI is the fully managed AI cloud for you to develop, train, and deploy AI models, with production-grade performance, cost efficiency, comprehensive ML tooling, and flexible GPU options backed by enterprise SLAs.

What Lepton offers you:

• Cloud Platform: A fully managed AI cloud platform that enables you to develop, train, and deploy models.

  • Serverless Endpoints: Directly use serverless endpoints as OpenAI-compatible APIs out of the box.
  • Dedicated Endpoints: Deploy dedicated endpoints that are fully customizable with your own models and configurations.
  • Dev Pods: Run interactive development sessions, including good old SSH, Jupyter notebooks, VS Code, etc., with managed GPUs in the cloud.
  • Batch Jobs: Run distributed training or batch processing jobs, with high performance interconnects and accelerated storage.
  • Platform Features: Access to serverless file systems, databases, network ingress and related tools with ease while you build your application.
  • Workspace Settings: Configure your workspace settings, including plan, members, billing, token, secret, registry, etc.

• Lepton for enterprise: Dedicated node groups for enterprise users.

  • Managed Node Groups: Create node groups that are managed by Lepton, with high performance GPUs and storage.
  • Bring Your Own Machines: Bring your own machines to create a self-managed node group.

• Python SDK for AI applications: Build AI applications with our Python SDK.

  • Lepton AI lives on GitHub. Whenever you have a question, feel free to file an issue.

What's next?

If you are new to lepton, we recommend you start with the Quickstart guide.