Nvidia Releases Open-Source AI Model Nemotron 3.5 Lightning

Nvidia has released Nemotron 3.5 Lightning, a new open-source AI model with 30 billion parameters, marking a significant expansion of its open AI portfolio. The model, unveiled on August 11, 2026, is designed to run on a single graphics processing unit, making advanced AI capabilities more accessible to developers and organizations without extensive computing resources. This release is part of Nvidia's broader strategy to foster innovation and maintain its central role in the rapidly evolving artificial intelligence industry.

The Nemotron 3.5 Lightning model is engineered for high efficiency and speed, particularly for autonomous AI agents that perform tasks in the background. According to Nvidia, it delivers four times the throughput in its class compared to its Nemotron 3 Super model and offers a 30 percent speed increase in completing agentic AI tasks over comparable models. The model leverages Nvidia's hybrid Mamba transformer architecture and multitoken prediction technology. It has been tested by companies including CrowdStrike, CodeRabbit, and Harvey for applications such as code review, tool use, and security alert monitoring.

Alongside the model, Nvidia introduced NeMo...