NVIDIA unveils GR00T 2 robot model and Isaac Lab 3.0
NVIDIA says it completed 40 of 50 evaluated tasks at least once without task-specific demonstrations, and reached 71% average success across 50 tasks with one demonstration each, rising to 95% with five. The company also released Isaac Lab 3.0, its open source robot-learning framework, which adds switching between the PhysX and Newton physics solvers, a reinforcement post-training library for foundation models, and Newton-based modeling of deformable objects such as cables and cloth. A Video to Data workflow and the EgoHand hand-motion model are also available. Foxconn, Siemens and Nexuni are testing GR00T 2.
Why it matters
NVIDIA is releasing the model and simulation stack openly, so robotics teams can train dexterous manipulation with far fewer task-specific demonstrations and rehearse delicate work in simulation before touching hardware.
Sources
- PrimaryNVIDIA
Written by AIToolMaps from the source above and checked against it sentence by sentence.