NVIDIA releases Nemotron 3.5 Bolt, a 4GB open model for local voice agents

NVIDIA released Nemotron 3.5 Bolt E4B 4GB, an open small language model post-trained from Gemma 4 E4B. It targets local voice assistants, game characters and robots, and is sized to run in about 4GB of VRAM on RTX and Jetson hardware.

Why it matters: Small on-device models that fit consumer GPUs and edge robotics boards push voice agents away from cloud APIs, cutting latency and cost. It also shows NVIDIA using open third-party bases to pull developers toward its own RTX and Jetson hardware stack.