Google DeepMind launches Gemini Robotics 2.5 with 96% accuracy sorting Mercedes-Benz parts
GoogleDeepMind · x · 2026-11-10
Google DeepMind released Gemini Robotics 2.5, a physical AI model that learns hundreds of everyday skills, from folding laundry to autonomously sorting car parts on the factory floor.
Key facts
- In a proof of concept with Apptronik and Mercedes-Benz, the Apollo 2 robot used it to pick and organize parts for assembly with 96% accuracy.
- Dexterity approaching human level: tying a bow (deformable objects), pipetting liquids (delicate handling), serving food with tongs (tool use).
- Gemini 4 Argon helped build it, generating virtual practice spaces from a single camera view and reviewing its own failure replays—speeding up months of training.
- Safety: built-in hazard detection, steering away from humans, and jailbreak testing.
More from Embodied
- AI2 open-sources ToyBlocks to benchmark robotic world models by task-relevant fidelity — allen_ai · 2026-11-11
- OpenTac: open-source ecosystem turns any robot surface into a tactile skin — _krishna_murthy · 2026-11-11
- NVIDIA launches open Alpamayo 2 Edge for real-time in-vehicle AI reasoning — nvidia · 2026-11-11
- NVIDIA open-sources GR00T 2 robot foundation model with 100K hours of training data — DrJimFan · 2026-11-11
- Stanford DroneHacks: 100 students fly GPS-denied PX4 drones on stereo cameras in 48 hours — mrpollo · 2026-11-10
- UT Austin open-sources OopsieVerse, a damage-aware safety benchmark for household robots — DynamicWebPaige · 2026-11-10