Google DeepMind Unveils Gemini Robotics 2 to Power Robots of All Shapes

The Decoder · rss · 2026-08-01

Google DeepMind has introduced Gemini Robotics 2, its most advanced vision-language-action (VLA) model to date.

The model is designed to control robots of all shapes, ranging from tabletop arms to full-body humanoids. Additionally, the newly added Gemini Robotics ER 2 features a higher-level reasoning layer specifically for handling complex robotics tasks.

Original post →

More from Embodied

Embodied channel →