Nvidia releases Nemotron 3.5 Lightning, an open 30B-parameter MoE model that it says delivers up to 4x faster output speeds, and an agentic AI model router (Kyt Dotson/SiliconANGLE)
Why it matters — For engineers building AI applications, an open MoE model with faster inference could lower latency and cost, while the agentic router hints at simpler orchestration across models. However, the speed claim is unverified and details on the router are minimal, so adoption should follow benchmarking.
↗