Nvidia Launches Multimodal Nemotron 3 Nano Omni Model

Nvidia has launched Nemotron 3 Nano Omni, a 30-billion-parameter AI model unifying text, vision and speech in a single architecture. Using a mixture-of-experts design, it delivers up to nine times faster throughput than competing open omni models, enabling real-time screen interpretation, document understanding and voice interaction for agentic AI applications. The model is compact enough to run on high-end consumer hardware and enterprise cloud deployments, and is designed to work alongside other Nvidia Nemotron models. It is now available on Hugging Face, OpenRouter and build.nvidia.com as an Nvidia NIM microservice.