NVIDIA Launches Nemotron 3 Nano Omni Model, Unifying Vision, Audio and Language for up to 9x More Efficient AI Agents

NVIDIA has launched the Nemotron 3 Nano Omni, an open multimodal model that integrates vision, speech, and language into one system. This advanced model enables agents to deliver faster, smarter responses, excelling on six leaderboards for complex document intelligence and audio-visual understanding. The Nemotron 3 Nano Omni functions as the 'eyes and ears' in agent systems, offering full deployment flexibility. It features a 30B-A3B hybrid architecture with Conv3D and EVS, achieving leading accuracy and nine times the throughput of other open models, enhancing scalability and responsiveness.