NVIDIA Releases Nemotron 3.5 Super VL and Two Bolt Models for Local Agents

NVIDIA has released the open 120-billion-parameter Nemotron 3.5 Super VL and two Nemotron 3.5 Bolt models aimed at agent deployments ranging from data centers to local devices. Super VL and Bolt 74B-A9B are available now, while Bolt E4B is scheduled for Oct. 23.

Super VL targets multimodal agent tasks including instruction following, tool use, multistep business workflows and visual analysis. NVIDIA says its accuracy is comparable to that of the company’s largest reasoning model at roughly one-fifth of the parameter count.

Bolt models target local deployment

Nemotron 3.5 Bolt 74B-A9B 24GB uses native 2-bit quantization to fit within 24GB of memory and run on NVIDIA RTX systems and DGX Spark alongside other applications. It is designed for local tasks ranging from web research to writing and debugging code without sending workloads to the cloud.

Bolt E4B 4GB is designed for GeForce RTX, Jetson and RTX/DGX Spark. NVIDIA positions it for fast voice assistants, interactive characters and embedded interfaces with multistep agent capabilities.

Super VL training and deployment

Super VL ships with open weights, datasets and training recipes. NVIDIA says it was trained on about 26 trillion tokens, then post-trained on roughly 3.8 million samples with reinforcement learning across dozens of environments.

Organizations can post-train the model on their own data, tools, policies and workflows. NVIDIA cited deployments and customization efforts involving financial analysis, cybersecurity, coding, medical claims, scientific search and other specialized tasks.

The model handles text, images, webpages, PDFs, tables and charts. NVIDIA also says deployment on its Dynamo platform can serve up to eight times more users with 85% lower latency.

Super VL is available through Hugging Face, OpenRouter, ModelScope and NVIDIA NIM. Bolt 74B-A9B is available through Hugging Face and supports llama.cpp, Ollama, LM Studio, Unsloth and vLLM; it is also being prepared for NVIDIA RTX Spark laptops arriving Oct. 16.