NVIDIA has expanded its Nemotron 3 model family with Nemotron 3.5 Lightning, a model the company says is designed to improve the speed and efficiency of long-running agentic artificial intelligence workloads. The announcement also highlights NeMo Switchyard, NVIDIA’s platform for supporting these applications. The company said the release responds to growing demand for open AI models that give organizations greater control over where systems run, how they are deployed and how they are updated. NVIDIA described Nemotron 3.5 Lightning as the highest-efficiency model in its class for agentic AI workloads. The announcement reflects the broader shift from conversational chatbots toward autonomous AI agents capable of carrying out multistep tasks. NVIDIA published details about the models and related tools through its newsroom on Aug. 11, 2026. Post navigation Samsung Brings Gaming Showcase to Gamescom Galaxy Buds Gain Hearing Aid Clearance