// AI news · 2026-11-10 · Google DeepMind Blog
Gemini Robotics 2.5 Enhances Real-World Dexterity
Google DeepMind introduces Gemini Robotics 2.5, a vision-language-action model that enhances dexterity, reliability, and safety in real-world robotics.
Published 2026-11-10 03:51 (Beijing time)Source Google DeepMind BlogAuto-summarised, fact-checked against source
01Summary
It performs across over 200 tasks with up to 100% higher success rates and 20–30% faster execution. Tested on Mercedes-Benz production lines, it achieved 96% success in part picking and packing. The model supports multiple robot platforms and includes real-time safety steering for physical environments.
Summarised from the original article by an open-source model, then checked automatically: every number and English name must appear in the source. The original article is authoritative.
02Key points
- Gemini Robotics 2.5 enables over 200 distinct tasks with up to 100% higher success rates.
- Achieved 96% success rate in automotive part kitting on production lines.
- Features real-time physical hazard avoidance through inference-time reasoning.
- Works natively with Apollo 2 and Franka Duo robot platforms.
03Source
Gemini Robotics 2.5 accelerates real-world general dexterity
Google DeepMind Blog · published 2026-11-10 03:51 Beijing time
05Related news
- Same source · Google DeepMind Formally Verifies Signal's Crypto with AI2026-11-10 · Google DeepMind Blog
- Same source · AlphaGenome 2 Predicts Variant Effects in Cell Types2026-11-11 · Google DeepMind Blog
- Same topic · Legacy Software Hinders Enterprise AI Adoption2026-11-10 · Google Cloud AI/ML
- Same topic · Gemini 4 Carbon Available on AI Gateway2026-11-10 · Vercel Changelog
- Same day · Databricks Expands Sovereign Cloud in Europe2026-11-10 · Databricks Blog
From AIMap AI news: collected from 1,100+ official and media sources. Summaries only restate the original article; follow the link for the full story. All stories of 2026-11-10 · Archive by date · All AI news · Today