NVIDIA introduced Project GR00T, a universal base model for humanoid robots and the new Jetson Thor computer on the NVIDIA Thor system-on-chip (SoC).

The company also presented significant updates to the NVIDIA Isaac robotics platform, including basic generative AI models, simulation tools, and AI workflow infrastructure.

“Building foundation models for general humanoid robots is one of the most exciting problems to solve in AI today,” said Jensen Huang, founder and CEO of NVIDIA. “The enabling technologies are coming together for leading roboticists around the world to take giant leaps towards artificial general robotics.”

Robots based on GR00T, which stands for Generalist Robot 00 Technology, will be able to understand speech and imitate movements by observing human actions. In his report, Huang demonstrated several such robots.

Jetson Thor was created as a new computing platform capable of performing complex tasks and interacting safely with humans and machines. It has a modular architecture optimized for performance, power, and size.

The SoC includes a next-generation GPU based on NVIDIA Blackwell’s NVIDIA Transformer Engine (TE) architecture, which delivers 800 teraflops of 8-bit floating-point AI performance to run multimodal generative AI models such as GR00T.

The integrated functional safety processor, high-performance CPU cluster, and 100 GB of Ethernet bandwidth greatly simplify design and integration.

NVIDIA provides a comprehensive AI platform for leading humanoid robotics companies such as 1X Technologies, Agility Robotics, Apptronik, Boston Dynamics, Figure AI, Fourier Intelligence, Sanctuary AI, Unitree Robotics, and XPENG Robotics, among others.

“We are at an inflection point in history, with human-centric robots like Digit poised to change labor forever. Modern AI will accelerate development, paving the way for robots like Digit to help people in all aspects of daily life,” said Jonathan Hurst, cofounder and chief robot officer at Agility Robotics. “We’re excited to partner with NVIDIA to invest in the computing, simulation tools, machine learning environments and other necessary infrastructure to enable the dream of robots being a part of daily life.”

NVIDIA also announced Isaac Manipulator and Isaac Perceptor, a collection of pre-trained models, libraries, and reference hardware for robotics.

Isaac Manipulator offers modular AI capabilities for robotic arms with a collection of GPU-accelerated base models and libraries.

The technology accelerates trajectory planning by 80 times. Among the first ecosystem partners: Yaskawa, Universal Robots, Teradyne, PickNik Robotics, Solomon, READY Robotics, and Franka Robotics.

The Isaac Perceptor provides multi-camera 3D volumetric vision that is increasingly being used in autonomous mobile robots used in manufacturing and warehouse operations to improve worker efficiency and safety, as well as reduce errors and costs.

Among the first to implement this technology are ArcBest, BYD, and the KION Group, which are striving to achieve new levels of autonomy in material handling operations, etc.