DeepMind Unveils Gemini Robotics 2 for Whole-Body Humanoid Control
lukas_m_ziegler · x · 2026-07-31
@GoogleDeepMind released Gemini Robotics 2, a suite of three models working together to handle the practical needs of real robots.
- Whole-body Control: As a Vision-Language-Action (VLA) model, it controls the entire humanoid body as a single integrated system at scale for the first time, rather than controlling arms and legs separately. It reasons through whole-body movements while walking, crouching, and bending.
- Advanced Dexterity: Demonstrates high-level manipulation skills on 22-degree-of-freedom hands.
- Multi-robot Collaboration: A humanoid and a bi-arm robot with different embodiments can share understanding to collaborate on the same task, communicating and dividing work based on spatial reachability.
- Embodied Reasoning: Gemini Robotics ER 2 acts as the high-level brain, planning multi-step tasks, conversing with humans, handling uncertainty, and requesting human intervention when unsure.
Related event: Google Unveils Gemini Robotics 2 and ER 2 for Full-Body Robot Intelligence(71 posts)→
More from Embodied
- NVIDIA launches RTX Spark for local AI on Windows PCs — ostrisai · 2026-10-16
- Transforming quadruped robot runs on real hardware: per-pose RL policies in Isaac switched on morph — _Stocko_ · 2026-10-16
- Foldable iPhone Duo reportedly sells out launch-day inventory, some deliveries slip to 2027 — Polymarket · 2026-10-16
- KinoGen synthesizes robot loco-manipulation reference motions directly via kinematically grounded video generation — ChongZzZhang · 2026-10-16
- Science Robotics paper introduces F³, predictive event-based vision beyond optical flow — CSProfKGD · 2026-10-16
- iPhone Duo foldable appears on South African retailer iStore at R47,999 in apparent leak — shauntrennery · 2026-10-16