NVIDIA has expanded its Nemotron 3 model family with the addition of Nemotron 3.5 Lightning, which the company describes as the highest-efficiency model in its class for long-running agentic AI workloads. The release accompanies NeMo Switchyard and follows an earlier Nemotron release. NVIDIA frames the launch around a broader shift from chatbot-style applications toward autonomous agents, and positions open models as a way to give users control over where AI runs and how it is deployed and evolved.
Why it matters
The announcement reflects growing attention to agentic AI, where models operate over extended tasks rather than single exchanges. NVIDIA emphasizes efficiency for these longer workloads and the deployment flexibility that open models can provide.
Who should care
Developers and organizations building or running autonomous AI agents may find the efficiency focus and deployment control relevant to their needs.