Gemini Robotics 2 brings whole-body intelligence to humanoid robots
Google DeepMind launched Gemini Robotics 2, a three-model physical AI stack: Gemini Robotics 2 (vision-language-action for full humanoid control from feet to fingertips), Gemini Robotics ER 2 (embodied reasoning agent for multi-step planning and multi-robot collaboration), and Gemini Robotics On-Device 2 (efficient local VLA that adapts to new robot embodiments in a few hours). ER 2 is available via Google AI Studio and private preview on Gemini Enterprise Agent Platform; VLA and On-Device models ship to early-access partners, with a new ASIMOV-Agentic safety benchmark.




