NVIDIA expands Nemotron family with 120B Super VL and edge-focused Bolt models

The Nemotron family grows

NVIDIA added three new members to its open-weight Nemotron model family. The new flagship is Nemotron 3.5 Super VL, a 120-billion-parameter model. According to the company, it delivers the highest accuracy and fastest task completion in its class for multimodal agentic tasks — instruction following, tool use, multistep business workflows and visual understanding.

NVIDIA says Super VL is "5x faster than the next best" model in its class and reaches accuracy comparable to the company's largest reasoning model at roughly one-fifth the parameter count. On the Artificial Analysis Intelligence Index leaderboard, it is said to achieve up to 5x faster throughput than other models in its class.

Two Bolt models for the edge

• Nemotron 3.5 Bolt 74B-A9B 24GB: Designed to run everyday agentic tasks locally on NVIDIA RTX and DGX Spark. Native 2-bit quantization lets it fit in 24GB of memory and run alongside other applications.
• Nemotron 3.5 Bolt E4B 4GB: Built for physical and conversational agents on GeForce RTX, Jetson and RTX/DGX Spark, targeting fast local voice assistants, interactive characters and embedded interfaces.

Super VL and Bolt 74B-A9B 24GB are available now; Bolt E4B 4GB arrives on Oct. 23.

Open models, your own infrastructure

NVIDIA stresses that, unlike closed frontier models, organizations can post-train Super VL on their own data, deploy it on their own infrastructure and oversee what they build. According to the company, the three models cover the full range of agent deployments, from large enterprise servers down to edge devices.

Who is customizing it?

NVIDIA says dozens of organizations are post-training Super VL on their own data, tools, policies and workflows. Examples include Scale AI's agents for Skydance and EY, AfterQuery's credit agreement analysis for Hebbia, AI21 and Alice's cybersecurity resilience work, Mercor's APEX-SWE coding benchmark and ModelScope's fine-tuning support through its ms-swift framework.