Dispatch
NVIDIA Launches Nemotron 3 Nano Omni — Unified Open Multimodal Model With 9x Efficiency Gain
NVIDIA released Nemotron 3 Nano Omni, a 30B-parameter (3B active) open multimodal model that unifies video, audio, image, and text reasoning in a single hybrid MoE architecture. It delivers 9x higher throughput than comparable open omni models and tops six leaderboards for document intelligence, video, and audio understanding. Available day-0 on HuggingFace, OpenRouter, build.nvidia.com, and 25+ partner platforms — early adopters include Palantir, Foxconn, and DocuSign.
Source
↳ Follow the thread