Amazon SageMaker AI Adds Serverless Customization for NVIDIA Nemotron 3.5 Lightning
Amazon SageMaker AI now offers serverless model customization for NVIDIA Nemotron 3.5 Lightning, supporting supervised fine-tuning (SFT), Direct Preference Optimization (DPO), and reinforcement fine-tuning (RFT). This capability allows organizations to tailor foundation models with proprietary data to achieve frontier-model quality with smaller, cost-effective, and low-latency models. With serverless customization, SageMaker AI manages infrastructure, enabling users to focus on data and evaluation rather than cluster management. The feature is available in US East (N. Virginia), US West (Oregon), Asia Pacific (Tokyo), and Europe (Ireland) regions.
Features (1) ›
- Serverless Customization for NVIDIA Nemotron 3.5 Lightning on SageMaker AI
Amazon SageMaker AI now supports serverless model customization for the NVIDIA Nemotron 3.5 Lightning model, utilizing supervised fine-tuning (SFT), Direct Preference Optimization (DPO), and reinforcement fine-tuning (RFT). This enables users to adapt foundation models with proprietary data to match frontier-model quality on specific tasks, reducing costs and latency, with SageMaker AI handling all infrastructure provisioning and training orchestration.
https://aws.amazon.com/about-aws/whats-new/2026/09/amazon-sagemaker-ft-nemotron-3-5-lightning/
Related releases
- Discover and Govern Snowflake Data with SageMaker Unified Studio AWS Big Data Blog ·
- AWS Step Functions automatically integrates new AWS services AWS What's New ·
- Connecting SageMaker Unified Studio to Power BI with IAM-Based Domains AWS Big Data Blog ·
- Direct Power BI connectivity to SageMaker Unified Studio via Athena ODBC driver AWS Big Data Blog ·
- Amazon SageMaker Adds Instance Preference Lists for Training and Processing Jobs AWS What's New ·
- Amazon SageMaker Unified Studio Adds ODBC Support for Power BI and Analytics Tools AWS What's New ·