Just when you thought the humanoid robot space couldn’t get more crowded, Unitree Robotics has decided to give its H2 model a serious cognitive boost. The newly unveiled Unitree H2 Plus is less of an incremental update and more of a full-blown brain transplant, thanks to its integration of NVIDIA’s monstrously powerful Jetson Thor superchip. This move firmly plants the H2 Plus as a high-end development platform for the next generation of AI-driven robotics.
The H2 Plus itself is no slouch in the hardware department. Standing at 1.82 meters (nearly 6 feet) and weighing around 70kg, it’s a substantial piece of kit. The robot boasts 31 degrees of freedom in its main body, with impressive peak torque figures of 360 N·m in the legs and 120 N·m in the arms. But the real party trick is the optional Dual SharpaWave Tactile Five-Finger Dexterous Hands, which add another 22 degrees of freedom per hand, bringing the grand total to a dizzying 75 DOF. Each fingertip on these advanced manipulators contains over 1,000 tactile pixels, giving the robot a sense of touch with a force sensitivity down to 0.005N.
The star of the show, however, is the NVIDIA hardware and software stack. The Jetson Thor system-on-module provides a staggering 2,070 TFLOPS of FP4 AI performance, powered by a Blackwell architecture GPU and a 14-core Arm CPU. This onboard supercomputer is designed to run the entire NVIDIA Isaac platform, including the GR00T foundation models for general-purpose robotic tasks. This means developers can leverage NVIDIA’s extensive tools for simulation (Isaac Sim), data capture (Isaac TeleOp), and deploying trained policies directly onto the robot.
Why is this important?
Unitree, a company that made its name with surprisingly affordable and agile quadrupeds, is making a clear and aggressive push into the high-performance humanoid arena. By partnering with NVIDIA to create an open reference design, Unitree is not just selling a robot; it’s selling a complete, ready-to-research platform. This dramatically lowers the barrier for academic and commercial labs to start working with cutting-edge hardware that’s pre-integrated with a powerful AI software stack. Instead of spending months integrating components, researchers can get straight to developing skills and behaviors using GR00T foundation models. The H2 Plus is a direct challenge to more vertically integrated players, betting that an open, powerful, and accessible platform will accelerate the entire field.
