Google DeepMind introduces Gemini Robotics 2
Google DeepMind has unveiled Gemini Robotics 2, an advanced AI model designed to bring whole-body intelligence and fine-grained motor control to robotic systems. By integrating multimodal reasoning with real-time sensorimotor actuation, it equips robots to coordinate complex movements, perform delicate tactile tasks, and collaborate across multi-agent environments.
Gemini Robotics 2 marks a pivotal shift from localized manipulation toward holistic, real-time whole-body robotics powered by vision-language-action foundation models.
* Unifies spatial reasoning, tactile perception, and whole-body motor control within a single multimodal model architecture.
* Enables fine dexterity and multi-robot coordination required for complex, real-world task execution.
* Accelerates the general utility of humanoid robotics by replacing hand-crafted motion planning pipelines with end-to-end neural intelligence.
DISCOVERED
2h ago
2026-07-30
PUBLISHED
3h ago
2026-07-30
RELEVANCE
AUTHOR
ai2027