NVIDIA releases Nemotron 3.5 Super VL and two open Bolt models

NVIDIA 发布开源 Nemotron 3.5 Super VL 及 Bolt 两款端侧模型

NVIDIA released Nemotron 3.5 Super VL and two on-device Bolt models for agent deployments on enterprise servers, edge devices and local hardware. Super VL handles visual document understanding, tool use and multi-step workflows, with open weights, datasets and training recipes. Bolt 74B-A9B 24GB uses native 2-bit quantization and runs local agent tasks in 24GB of memory.

Why it matters: NVIDIA now offers open models for both enterprise servers and edge devices, so teams can pick by deployment target and tune agent tasks on their own data.