Manage Amazon SageMaker JumpStart foundation model access with private hubs

This article discusses how enterprise administrators can manage access to foundation models in Amazon SageMaker JumpStart using private hubs. It provides steps for administrators to create a private hub, curate a list of allowed models, configure access control, and share the private hub across accounts. It also shows how users can access and deploy models from the private hub using SageMaker Studio and the SageMaker Python SDK.

Specifically, the article covers:

• Solution overview of private hubs for managing foundation model access
• Prerequisites for using the SageMaker Python SDK
• Steps for administrators to create a private hub, curate models, and configure access control
• Steps for users to interact with allowlisted models in the private hub via SageMaker Studio and Python SDK
• Cross-account sharing of private hubs using AWS Resource Access Manager (RAM)
• Conclusion on benefits of private hubs for enterprise governance of foundation models