Google DeepMind Unveils Gemini Robotics 2
GoogleDeepMind · x · 2026-08-07
Google DeepMind introduces Gemini Robotics 2, an intelligence layer powering the next generation of adaptable robots. The system combines multimodal understanding with whole-body control, unlocking three core capabilities:
- Whole-body intelligence and dexterity: Robots can reason through movements to perform complex tasks like walking, crouching, stretching, and cleaning cluttered rooms.
- Multi-robot collaboration: Enables multiple robots to team up and finish jobs faster.
- On-device execution and adaptation: The model can run locally on-device and seamlessly adapt to entirely new robotic bodies in just a few hours.
More from Embodied
- OpenRoboto Launches on Bittensor to Build Open-Weight Robotics Models — markjeffrey · 2026-08-08
- Agriculture Bot Powered by XTR-0 Brain: Edge Computing Meets Robotics — mjdramstead · 2026-08-08
- Many winners expected in robotics, exciting and rewarding field — chris_j_paxton · 2026-08-08
- VRAM Tier Testing: 10GB VRAM + 32GB RAM Runs 35B Sparse Model — OutdatedMemeKing · 2026-08-08
- OpenAI's First Smart Speaker Expected in 2027, Priced Over $300 — The Decoder · 2026-08-08
- DIY Outdoor Utility Robot U-BOT: Testing 3D Printed Drive Components — _Stocko_ · 2026-08-08