Om AI has launched VLX, a series of models designed for real-time interaction with the physical world. Unlike traditional models that process video frames offline, VLX uses a novel "streaming multimodal" architecture for continuous, millisecond-level perception and action. The series includes VLX-Flow for ongoing environmental awareness, VLX-Seek for precise spatial localization, and VLX-Go for direct translation of visual understanding into robotic actions. AI
IMPACT Enables real-time, continuous perception and action for edge devices, potentially accelerating embodied AI development.
RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
