NVIDIA releases open Nemotron 3.5 Super VL: 120B params, 12B active, multimodal

yoavartzi · x · 2026-10-16

NVIDIA released Nemotron 3.5 Super VL, an open model with 120B total parameters and 12B active, combining reasoning and tool use with text, image, document, and video understanding for long-running AI agents. It claims strong multimodal reasoning and high output speed, is open and customizable, and is available on Hugging Face.

Related event: NVIDIA Open-Sources Nemotron 3.5 Vision-Language Model Family(30 posts)→

Original post →

More from Models

Models channel →