Sakhanda Wire
NVDA $154.00 -1.37% MSFT $506.28 -2.26% GOOGL $280.87 -1.00% META $620.34 -0.97% AMZN $256.88 -0.34%
← Back to the news

Google DeepMind reports 96% robot sorting success at Mercedes-Benz

Google DeepMind has introduced Gemini Robotics 2.5, a vision-language-action model that interprets visual data and instructions before converting them into robot movements. The company says it was evaluated on more than 200 tasks and delivered success rates 30% to 100% higher than its previous internal baselines, depending on the task, while completing work 20% to 30% faster.

In cooperation with Apptronik and Mercedes-Benz, the model was tested on active production lines in Kecskemét, Hungary, and Sindelfingen, Germany. Apptronik’s humanoid Apollo 2 sorted metal components from bulk containers and placed them in vehicle-specific tray sequences. Google DeepMind reported a 96% success rate. Gemini Robotics ER supervised the workflow, while Gemini Robotics 2.5 controlled the robot’s arms and hands. Apptronik CEO Jeff Cardenas said post-training produced substantial improvements within weeks and that the system moved into Mercedes production with limited performance loss.

Google DeepMind also reported improved safe task completion near obstacles and people. Its open-source ASIMOV-Jailbreak benchmark tests whether robotics models can be induced to perform dangerous actions. The company used Gemini 4 Argon to generate simulated training runs and digital twins, and said Gemini Robotics 2.5 is being rolled out to Agile, Apptronik, and Mercedes-Benz.

Written by Sakhanda Wire AI from reporting by The Decoder. Facts, quotes and full context are in the original article below.

Originally published by The Decoder on

Read the original on The Decoder ↗

The reporting and the image belong to The Decoder and are credited and linked above; the text on this page is Sakhanda Wire's own summary.

← Back to the news

More stories

All the latest news