NVIDIA launches Nemotron 3.5 Lightning and NeMo Switchyard for efficient agentic AI

NVIDIA is expanding its Nemotron 3 model family with the release of Nemotron 3.5 Lightning, a 30-billion-parameter mixture-of-experts model designed for specialized tasks within larger multi-agent systems. The model delivers up to 4x faster output speed and 30% faster agentic task completion compared to others in its class. It is fully customizable and can be post-trained with NVIDIA NeMo on an organization's domain data, tools, and workflows. The company also released NeMo Switchyard, an open-source library for model routing that automatically directs prompts to the most capable and efficient model for each step of an agent workflow. Internal benchmarks show NeMo Switchyard maintains frontier-level accuracy while reducing task completion cost to nearly a third of Opus 4.8 alone. Partners including CrowdStrike, Harvey, CodeRabbit, Lila Sciences, and Fastino Labs are customizing Nemotron 3.5 Lightning for their workloads, while Boomi, Cadence, Cognition, Kong, LangChain, LiteLLM, Nous Research, Ramp, Siemens, and Classmethod are integrating or evaluating NeMo Switchyard.