Nvidia released Nemotron 3.5 Super VL on October 15, 2026, saying it is built to run enterprise AI agents nonstop. It is the company's first major update to its Nemotron series since the release of Nemotron 3 Ultra in August 2026.
Nvidia reported the model scores 80.8 percent on the chart benchmark CharXiv, measured on a set of visual tasks. That compares with 63 percent for the older Nemotron 3 Nano Omni.
Nemotron 3.5 Super VL is built on a Mixture-of-Experts architecture and targets continuous background processing for text, images, documents, and videos. Availability begins with the release of the model, initially for enterprise users and developers.
"Super VL delivers up to five times the throughput of other open models of similar size on agent tasks," said Jonathan Kemper, a contributor at thedecoder. Nvidia claims the model's efficiency allows it to serve eight times as many users while cutting time to first token by 85 percent.
The announcement follows the release of the Nemotron Lightning model in August 2026. Nvidia framed the significance of Super VL as a step toward more efficient and scalable AI agent solutions.
Nvidia did not say how the model will perform against closed frontier models, and raised the question of whether its open approach will appeal to enterprises seeking full customization. The company said it will continue to release updates and optimizations for the model.
Source: thedecoder
