NVIDIA Nemotron 3.5 Lightning model is now available on Amazon SageMaker JumpStart

This article announces the availability of NVIDIA's Nemotron 3.5 Lightning model on Amazon SageMaker JumpStart for AWS customers.

• Hybrid Mixture-of-Experts architecture with 30B parameters and 3B active per forward pass
• Achieves up to 4x throughput (~410 tokens/sec) and 30% faster task completion than comparable models
• Handles up to 1M tokens of context via DFlash speculative decoding
• Engineered for persistent agents in personal assistants, financial processing, cybersecurity, and telecom
• Fully open-trained on open datasets, allowing enterprise post-training and deployment across edge, on-premises, or cloud
• Deploy in a few clicks via SageMaker JumpStart console or Python SDK

Nemotron 3.5 Lightning enables enterprises to rapidly deploy high-performance AI agents with complete ownership and customization capabilities.