Gemini Robotics ER 2
READER SUMMARYAn embodied-reasoning model that plans multi-step robot tasks, self-corrects, coordinates multiple robots, and is available through AI Studio and enterprise preview.
Why this model scores 67.9
ER 2 plans and tracks hundreds of decisions over several minutes, recovers from failures, coordinates robots, and has broader access than the VLA tiers. Safety constraint following, human-proximity stops, and retained human supervision lower control difficulty, while physical orchestration keeps autonomy and misuse above general non-embodied models of similar capability.
News tied to Gemini Robotics ER 2
The model score of 67.9 rates this model's risk profile. The overall Doom Index of 67.7 measures the accumulated evidence trajectory. These values answer different questions.
Each article's signed Doom Index movement is divided equally among the exact models named on that article. This prevents multi-model evidence from being counted in full on several model pages. The model's risk score itself never changes the overall index.
Google DeepMind releases Gemini Robotics 2 for whole-body and multi-robot tasks
Three new robotics models demonstrated whole-body humanoid control, dexterous manipulation, several-minute planning, self-correction, multi-robot collaboration, and rapid on-device adaptation, with reasoning access available in AI Studio.
- Full item movement
- +3.27
- Gemini Robotics ER 2 equal share
- +1.09
Model score history
- R1Doom Score 67.9
Initial source-backed profile for the exact embodied-reasoning release.
11 Aug 2026