NVIDIA Unveils 30B-Parameter Nemotron-3.5-Lightning Model

On August 11, NVIDIA launched the Nemotron-3.5-Lightning model, a new addition to the Nemotron-3 series featuring 30 billion parameters and built on a Mixture-of-Experts (MoE) architecture, specifically designed for specialized tasks within large-scale multi-agent systems. Compared to similar models, it delivers up to 4x faster output speeds and accelerates overall agent task completion by 30%. The model runs locally on NVIDIA RTX PCs, NVIDIA DGX Spark, NVIDIA DGX Station, and NVIDIA Jetson AI systems, and can scale to edge AI devices, NVIDIA RTX PRO workstations, data centers, and cloud environments.