NVIDIA has expanded its Nemotron family with new models aimed at both data center and edge use. The company released Nemotron 3.5 Super VL, an open 120-billion-parameter model described as delivering fast task completion and high accuracy for specialized always-on agents. Alongside it, NVIDIA introduced two edge models: Nemotron 3.5 Bolt 74B-A9B 24GB, optimized to balance intelligence and speed per GB for local agentic workloads, and Nemotron 3.5 Bolt E4B 4GB.
Why it matters
The additions extend NVIDIA’s open model lineup across deployment contexts, from data center-scale to desktop-class hardware. The Bolt models are positioned specifically for running agentic workloads locally, where efficiency per GB is a consideration.
Who should care
Developers and organizations building agent-based applications, particularly those seeking open models that can run on edge or local hardware, may find the new Super VL and Bolt releases relevant.