Model2h ago
Nvidia Nemotron 3.5 Lightning Hits SageMaker
The 30B-parameter Nemotron 3.5 Lightning now deploys via Amazon SageMaker JumpStart. It uses only 3B active parameters to run on a single GPU. This model delivers 4x higher throughput for high-volume agentic workloads compared to its class. Developers can now execute repetitive agent steps without configuring complex serving infrastructure or using frontier-scale hardware.
