Google DeepMind unveils Gemini Robotics 2 with whole-body control and multi-robot collaboration

On August 7, Google DeepMind launched Gemini Robotics 2, a new Vision-Language-Action (VLA) model designed to power a new generation of adaptive robots. The company highlighted three major breakthroughs: whole-body intelligent control, advanced dexterous manipulation, and multi-robot collaboration. A demonstration video was also released showcasing natural language interactions with a robot named Apollo, marking a significant step forward in embodied AI for physical coordination and teamwork.

Confirmed

Unconfirmed

Why It Matters

By combining multimodal understanding with whole-body control, Gemini Robotics 2 represents a major milestone in physical AI. Its capabilities in whole-body control and multi-robot collaboration are expected to drive the practical deployment of humanoids in complex environments, holding significant potential for industrial and service sectors.

2026-08-07 ~ 2026-08-07 · 5 related posts

Full story(6 episodes)→

Primary sources

3 near-duplicate retellings: gabriberton · GoogleDeepMind · GoogleDeepMind