Radar / AI robotics / NVIDIA GR00T 2
NVIDIA GR00T 2: a world-action robot model that learns from human video
NVIDIA's open frontier world-action model predicts future video and robot actions together, hitting 71% one-shot success across 50 tasks.
Why it matters
Trained on 100,000 hours of open-source human egocentric video, GR00T 2 transfers skills to robots with as few as one demonstration per task, and ships with Isaac Lab 3.0 for up to 2.5x faster simulation training.
What you could build with it
A robotics startup could train a manipulation policy on GR00T 2's human-video transfer, teaching a warehouse arm a new pick-and-place task from five human demonstrations instead of weeks of teleoperation data.
Does it hold up?
Too early to judge: code is listed as available in December and no weight license is published yet, though Foxconn and Siemens are already trialling it.
Built with NVIDIA GR00T 2
- NVIDIA Showcases GR00T 2 and Isaac Lab 3.0 at CoRL for Advanced Robot Trainingblockchain.news · CoRL unveiling details: GR00T 2 built on Cosmos, Isaac Lab 3.0 as the open-source robot-learning framework.
- Nvidia debuts GR00T 2 robotics model, 71% on 50 tasksaiweekly · Notes that NVIDIA gave no public release date or license for the GR00T 2 weights or the 100,000-hour dataset.
- Nvidia Unveils GR00T 2 Robot Brain and Isaac Lab 3.0, With Foxconn and Siemens Testing Industrial Rolloutbiggo · Trials underway at Foxconn, Siemens and Nexuni; partners include Skild AI, Bosch Research and UC Berkeley.
Learn more
First spotted on article: source.
More AI robotics
RuView: see people through walls with WiFi, no camerasOpen-source Rust system that turns ordinary WiFi signals into real-time pose estimation, vital-sign monitoring, and…robotics · JEV 0.76Runway Praxis-1: open-weight robot model that learned to act from videoRunway released Praxis-1, a 27B-parameter world-action model, with open weights, inference and fine-tuning code under…robotics · JEV 0.66RoboParty RP1: a full-stack open-source humanoid robot for embodied AI researchRoboParty's 1.3-metre humanoid, launched 16 October and back in coverage on 2 November, comes in 24-DoF standard and…robotics · JEV 0.62Gemini Robotics 2.5: DeepMind's most capable robot model, tested on a live Mercedes lineGoogle DeepMind introduced Gemini Robotics 2.5, its most capable vision-language-action model, reporting 30-100%…robotics · JEV 0.61T-Rex: a robot manipulation AI that reacts to fingertip touchPanasonic, UC Berkeley, NVIDIA and Stanford built T-Rex (Tactile-Reactive Dexterous Manipulation), a VLA model that…robotics · JEV 0.61Bear Robotics KINO-2: demonstration-to-skill robot-learning platformBear Robotics launched KINO-2 on Oct 21 at GTC Berlin: a robot-learning platform that turns human task demonstrations…robotics · JEV 0.59
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.