SageMaker AI now supports serverless fine-tuning for Nvidia Nemotron models

This article announces that Amazon SageMaker AI now supports serverless fine-tuning for Nvidia Nemotron 3 Nano models using supervised and reinforcement fine-tuning methods.

• Serverless customization for Nvidia Nemotron 3 Nano (30B parameters) now available
• Supports supervised fine-tuning (SFT) and reinforcement fine-tuning (RFT) methods
• Tailor models with proprietary data for domain-specific tasks and workflows
• AWS handles infrastructure provisioning and training orchestration automatically
• Pay only for resources used, no cluster management required
• Available in US East (N. Virginia), US West (Oregon), Asia Pacific (Tokyo), Europe (Ireland)
• Access via SageMaker Studio or Python SDK for programmatic control

SageMaker AI's serverless fine-tuning simplifies model customization by eliminating infrastructure management while enabling organizations to adapt Nemotron models to their specific needs.