Google DeepMind Ships Three Physical AI Models For Whole Body Control, Dexterity And Multi Robot Collaboration

Google DeepMind has unveiled Gemini Robotics 2, an intelligence layer designed for next-generation robotic systems. The release includes three distinct models focused on whole-body control for humanoids, embodied reasoning for task orchestration, and multi-robot collaboration capabilities.

This development marks a shift toward more sophisticated physical AI, moving beyond simple automation to complex dexterity and coordination. By integrating vision-language-action models, DeepMind aims to enhance how robots perceive and interact with dynamic environments.

The availability of these models could accelerate the deployment of humanoid robots in industrial and service settings. Industry observers note that improved embodied reasoning is critical for scaling robotic agents beyond controlled laboratory conditions.

This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.