GOOGLE'S ROBOTS SORTED MERCEDES PARTS ON A REAL FACTORY LINE — 96% RIGHT

Google DeepMind introduced Gemini Robotics 2.5 on November 10, its most capable vision-language-action model to date. In a proof of concept on active Mercedes-Benz production lines in Kecskemét, Hungary and Sindelfingen, Germany, Apptronik's Apollo 2 humanoids sorted automotive parts into kits with a 96% success rate — real factory work, not a lab demo.

DeepMind reports success rates 30% to 100% higher and tasks completed 20% to 30% faster than previous baselines across more than 200 tasks, and open-sourced an ASIMOV-Jailbreak benchmark for robot safety. One caveat: DeepMind published no trial count behind the 96%, so it is a vendor demonstration statistic, not an independently verified production number.