Google DeepMind embodied-intelligence model connecting perception, reasoning, dexterous manipulation, and whole-body robot control.
Features
- Unified vision-language-action modeling
- Whole-body motion intelligence
- Dexterous object manipulation
- Adaptation across robot forms
- Multi-robot collaboration
Use Cases
- General robotics research
- Industrial and service robots
- Dexterous grasping and manipulation
- Embodied-agent prototypes
- Multi-robot task orchestration
FAQ
Gemini Robotics 2 is Google DeepMind’s embodied-intelligence model for a new generation of robots. It combines visual-language reasoning with action capabilities so robots can understand open environments, perform dexterous manipulation, transfer skills across different robot forms, and collaborate on multi-robot tasks. Core capabilities include: Unified vision-language-action modeling, Whole-body motion intelligence, Dexterous object manipulation.
Common scenarios include: General robotics research, Industrial and service robots, Dexterous grasping and manipulation.
