NVIDIA's GR00T 2 Achieves 95% Success in Skill Transfer with Minimal Demonstrations
- Published
- Source
- InterestingEngineering.com
NVIDIA has unveiled GR00T 2, a cutting-edge robot foundation model capable of transferring human skills to robots with only five task demonstrations. This model achieved an impressive 95% average success rate across 50 tasks after a brief fine-tuning session, showcasing its ability to learn from extensive human video data and various robot interactions.
The significance of GR00T 2 lies in its innovative approach to skill transfer, allowing robots to learn from demonstrations performed on different devices. This cross-adaptation capability enables robots to execute tasks they have never practiced, enhancing their versatility and efficiency in real-world applications. The model's ability to predict future movements and interactions further solidifies its potential in the robotics field.
NVIDIA's GR00T 2 is designed to work with various robot bodies, including humanoids and dexterous hands. Future developments may focus on expanding its capabilities and refining its learning processes, as the technology continues to evolve and adapt to new challenges in robotics and automation.
Original report
NVIDIA’s GR00T 2 transfers human skills to robots with just five task demonstrations
InterestingEngineering.com · Neetika WalterThis briefing is an independently written summary based on publicly available reporting and is provided for industry information and news discovery. The original report and source publication are credited and linked where applicable. RobotToday does not claim ownership of third-party source material.
Rights concerns? Contact [email protected] with the relevant URL and details. We will review the matter and take appropriate action where warranted.
More related news
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
-
10
RobotToday Weekly November 2 – 6, 2026 RobotToday · 3 days ago
- 11
- 12
Related Suppliers
NVIDIA Robotics
United States
NVIDIA robotics division delivering Isaac simulation and training tools, Jetson edge compute modules, and Omniverse digital twin frameworks for physical AI.
Deci AI (Acquired by NVIDIA)
Israel
Israeli deep learning optimization startup acquired by NVIDIA in May 2024 for ~$300M; known for AutoNAC neural architecture search technology.
Enactic, Inc.
Japan
Tokyo startup behind OpenArm, a fully open-source 7-DOF bimanual humanoid arm for physical AI research, and Ena, a humanoid for care facilities.
AeiROBOT 에이로봇
Korea, Republic of (South)
Ansan, South Korea humanoid robotics startup spun off Hanyang University ERICA in 2018; builds the bipedal ALICE robot series.
Simple AI 深朴智能
China
Beijing embodied-AI startup building living-space service robots; its HiFi-UMI robot-free capture rig shipped with an open 2,000-hour dataset.
MindOn
Embodied AI startup founded 2025 by an ex-Tencent Robotics X researcher; its Mind-0 and Mind-1 models run humanoid and dual-arm robots.
Yamaha Agriculture, Inc.
United States
Yamaha Agriculture is Yamaha Motor's US-based farm unit, formed in 2025, selling the Prospr autonomous hybrid orchard vehicle and FAZER R spray helicopters.
Yamaha Motor Co Ltd (Robotics Division)
Japan
Yamaha Motor's robotics business in Iwata, Japan: SCARA, single-axis and Cartesian robots, SMT mounters, and bonders via Yamaha Robotics Co., Ltd.
TeknTrash Robotics
United Kingdom
London start-up deploying ALPHA, a wheeled dual-arm humanoid trained on VR-captured worker data to sort waste in recycling plants under a RaaS model.
Lili-o Inc.
Germany
Early-stage robot software startup building a hardware-agnostic deterministic execution engine that teaches humanoids skills from one human demonstration.
CosmicBrain AI
Physical AI firm scoping supervised humanoid pilots for data-center rack inspection and parts runs, and publishing the Cosmic 0.5 human-to-humanoid skill stack.
Maxinsights
United States
San Francisco vendor, founded 2024, of annotated egocentric video, hand/body mocap and tactile data for training robot VLA and world action models.
Robotics needs a service framework.
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
Join the RobotToday community on LinkedIn
Daily robotics news, in-depth analysis, conference highlights, and discussions with professionals worldwide.
