NVIDIA Nemotron 3.5 Lightning and NeMo Switchyard Deliver Faster, Smarter, More Efficient Agentic AI
Summary
NVIDIA announces Nemotron 3.5 Lightning, a high-efficiency 30B mixture-of-experts model for long-running agentic AI workloads, and NeMo Switchyard, an open-source routing library for directing prompts to the most capable models. The release emphasizes on-premises and edge deployment for privacy and cost efficiency, with benchmarks showing faster task completion and cross-model routing across a mixed model ecosystem.