Google Unveils Gemini Robotics 2 and ER 2 for Full-Body Robot Intelligence

Google DeepMind officially released the next-generation vision-language-action (VLA) model Gemini Robotics 2 and the embodied reasoning model Gemini Robotics ER 2. These models break the limitations of traditional pre-programmed or teleoperated robots, endowing humanoid robots with full-body intelligent control, advanced dexterous manipulation, and cross-embodiment collaboration, marking a transition from tabletop tasks to full-body intelligence in real human environments.

Confirmed

Why it matters

2026-07-30 ~ 2026-08-01 · 71 related posts

Full story(6 episodes)→

Primary sources

29 near-duplicate retellings: GoogleDeepMind · GoogleDeepMind · GoogleAI · OfficialLoganK · fofrAI · SongShuran · Matt Wolfe · osanseviero · lukas_m_ziegler · breadli428 · CyberRobooo · CyberRobooo · DynamicWebPaige · DynamicWebPaige · rseroter · DynamicWebPaige · DynamicWebPaige · soumitrashukla9 · keerthanpg · GoogleDeepMind · GoogleDeepMind · bousmalis · heiga_zen · CSProfKGD · jason_mayes · 量子位 · aigclink · keerthanpg · keerthanpg