# NVIDIA Nemotron 3.5 Lightning and NeMo Switchyard Deliver Faster, Smarter, More Efficient Agentic AI

Nemotron 3.5 Lightning is a 30B-parameter MoE model with up to 4x faster output; Switchyard cuts agent task costs to nearly a third.

- Nemotron 3.5 Lightning is available on Hugging Face, ModelScope, OpenRouter, and build.nvidia.com as an NVIDIA NIM microservice.
- It runs locally on RTX PCs, DGX Spark, DGX Station, and Jetson, or across data centers and cloud.
- NeMo Switchyard is open source on GitHub and integrates with partners like Boomi, Cadence, Cognition, LangChain, and LiteLLM.
- NVIDIA also released Nemotron-RL-Agentic-Terminal-Pivot, a reinforcement learning dataset for coding agents.

## Why it matters

Enterprises building agentic AI can now deploy a smaller open model and a routing library to cut costs and latency while maintaining accuracy.

## Sources

- [NVIDIA AI: NVIDIA Nemotron 3.5 Lightning and NeMo Switchyard Deliver Faster, Smarter, More Efficient Agentic AI](https://blogs.nvidia.com/blog/nemotron-lightning-switchyard-rtx-dgx)

---

Summarized by dstilled on 2026-08-11. https://dstilled.ai/story/26c03d20-aee3-4f9a-a44c-5c511fbe1564