Best-in-class open omni-modal reasoning model delivers the highest efficiency and accuracy to power agentic workflows such as computer use, document intelligence and audio-video reasoning.