Code & Chain · Signal Desk

Gemini Robotics 2 Achieves Full-Body Robot Intelligence, ER 2 Model Opens, Efficient Edge Version Ready

Original sourceGoogle DeepMindAdditional: blog.google

Summary

Google DeepMind has released Gemini Robotics 2, including three models: Gemini Robotics 2 (a vision-language-action model controlling full humanoid and dual-arm robots for dexterous manipulation), Gemini Robotics ER 2 (a vision-language model that can reason, plan multi-step tasks, and support mult…

Key points

  • Offers developers a complete robot intelligence stack from cloud reasoning to fully offline deployment, accelerating multi-agent robot application development.
  • The first universal robot intelligence solution simultaneously supporting multi-robot collaboration and rapid adaptation to new embodiments, poised to speed industrial and service robot deployment.
  • Robot developers can use ER 2 to quickly build understanding and planning capabilities, and deploy low-latency, offline control systems with the On-Device version, significantly shortening time-to-market.

Editorial note

This page is Code & Chain's editorial summary of public sources. It may be prepared with AI assistance and published through an automated workflow. Refer to the original sources; this content is not investment, legal, or tax advice.