Gemini Robotics ER 2 gives robots a higher-level reasoning layer

Google’s embodied-reasoning model plans physical tasks, follows video progress and can coordinate multiple robots.

Google DeepMind introduced Gemini Robotics ER 2 as a high-level brain for physical agents. It understands the environment, plans multi-step tasks, monitors video feedback and hands motor execution to lower-level vision-language-action models.

The model is available in public preview through the Gemini API and Google AI Studio. Its progress tracking and multi-robot coordination matter for robotics workflows where success must be verified, not merely predicted.

View original source