Nemotron 3.5 Super VL

Nemotron 3.5 Super VL is NVIDIA's vision-language model with 120 billion total parameters and 12 billion active per token. Its architecture combines Mamba-2, transformer attention, mixture-of-experts layers, and a C-RADIO vision encoder. The model accepts text, images, and video and produces text, supporting reasoning, coding, tool use, document understanding, and video understanding. Native Multi-Token Prediction enables speculative decoding. Sciforium's on-demand offering uses FP8 weights on MI325X GPUs with a 1,048,576-token context window.