NVIDIA Nemotron 3.5 Lightning Released on Amazon SageMaker JumpStart for High-Throughput AI

NVIDIA's Nemotron 3.5 Lightning, optimized for persistent agents and enterprise automation, became available on SageMaker JumpStart, offering fast processing and reduced task completion time.

By Noah Ellis · Editorial pseudonym · 2026-08-11 · 1 min read

AI-ASSISTED REPORTING · RETROSPECTIVE ARCHIVE

Retrospective edition: 2026-08-11. Source published: 2026-08-11. Added to this archive: 2026-11-03. This briefing was prepared later from the linked announcement; it was not published on this website on the historical edition date.

Nemotron 3.5 Lightning, a 30-billion-parameter model with only 3 billion active parameters per forward pass, leverages a hybrid Mixture-of-Experts architecture to achieve up to four times the throughput of competitors at roughly 410 tokens per second. This acceleration enables more rapid task execution in AI-powered assistants and automation across finance, cybersecurity, and telecommunications sectors.

This open-trained model supports long-context (up to one million tokens) processing via DFlash speculative decoding and integrates with popular agent frameworks. Full control over retraining and deployment enables enterprises to tailor the model to their requirements on edge, on-premises, or cloud systems using SageMaker JumpStart.

Source & further reading

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

Source published 2026-08-11. This report was prepared with AI assistance from the linked primary-source material. Claims and limitations should be checked against the original announcement or research.