Explore / Case study / Gemini Robotics ER 2

Gemini Robotics ER 2

Demo only

High-level spatial reasoning, task planning and multi-robot coordination from Google DeepMind.

by admin · listed 5 days ago

⚡ What System1 does here

The model interprets the environment, plans steps, uses tools and tracks progress. Lower-level vision-language-action models or other execution systems handle physical movements. This is a Gemini model, not a verified Jev integration.

About

Google DeepMind’s high-level embodied reasoning model for spatial understanding, task planning and multi-robot coordination. At verification, the official page lists ER 2 in public preview through Google AI Studio and the Gemini API. Evidence and limits: Public preview still requires compatible control systems, tools and hardware integration. The model card cautions about production use and excludes safety-critical uses where failure could cause harm or damage. We did not call the model or operate a robot. Original project images and demonstrations (with attribution):

Official multi-robot demonstration poster ↗

Official multi-robot coordination demonstration ↗

Official spatial understanding demonstration ↗

Sources: https://deepmind.google/models/gemini-robotics/embodied-reasoning/ https://deepmind.google/models/model-cards/gemini-robotics-er-2/

🧪 Community test reports

0 reports · 0 passed on the real model

No one has tested this yet. Be the first.

💬 Discussions

New post →

No discussions yet.