NVIDIA Launches Nemotron 3.5 AI Models for Enterprises and Desktops

16 Oct 2026, 2:18 pm · 2d ago · 1 min read · Analytics India Magazine

NVIDIA expanded its open Nemotron AI family by unveiling three new models designed for enterprise tasks, local AI agents, and conversational workloads across data centres and edge hardware. The lineup features Nemotron 3.5 Super VL, a 120-billion-parameter multimodal model trained on 26 trillion tokens with high-throughput visual understanding, alongside the lightweight Nemotron 3.5 Bolt series. Bolt 74B-A9B uses 2-bit quantisation to run locally on 24GB memory systems like NVIDIA RTX workstations, while Bolt E4B releases on October 23, providing flexible on-premise AI deployment options.