NVIDIA Releases Nemotron 3.5 Super VL and Bolt Models

NVIDIA released Nemotron 3.5 Super VL and two Bolt models for AI agents, as detailed in a company blog post. Super VL is an open model with 120 billion parameters that handles instruction following, tool use, visual documents and multistep workflows.

NVIDIA says Super VL completes tasks up to five times faster than comparable open models. Organizations can further train it with their own data and deploy it on their own infrastructure. The company provides open weights, datasets and training recipes, while NVIDIA NIM packages the model as a microservice.

Nemotron 3.5 Bolt 74B-A9B uses two bit quantization to fit within 24GB of memory for local work on RTX systems and DGX Spark. Super VL and Bolt 74B-A9B are available now, while the 4GB Bolt E4B model will arrive on October 23 for GeForce RTX, Jetson and DGX Spark systems.