Nemotron 3.5 Super VL: an open model that reads documents, images and video
On October 15, 2026, NVIDIA released Nemotron 3.5 Super VL, with 120B parameters and 12B active, open weights, data and recipes, text, image and video input and up to 1 million tokens of context.
According to the model page, the full-precision version runs on 4 H100 or 2 GB200, and there is also an NVFP4 quantized version. The license is OpenMDW-1.1. NVIDIA says the model was trained on about 26 trillion tokens and targets specialized agents, document and video understanding, without audio support.
For organizations with many PDFs, forms, charts or videos, an open model of this size can run on their own infrastructure. It is worth measuring quality on real documents and the GPU cost before deciding.