NVIDIA released Nemotron 3 Nano Omni, a 30-billion-parameter open-source AI model. This system processes text, images, video, and audio within a single unified architecture.
The model targets AI agent applications to reduce latency and improve performance. It achieves higher throughput than competing open multimodal models across industry benchmarks for document and video understanding.
NVIDIA released the model with open weights to encourage customization and challenge closed-source alternatives. Early enterprise adopters include Palantir, Foxconn, and Dell.