Unitree has introduced H2 Plus, the first humanoid robot reference design built on NVIDIA Isaac GR00T, combining a human-scale robot body, tactile five-finger hands (the “body”), onboard Jetson Thor compute, and Isaac GR00T software and workflows (the “brain”). The design gives research teams a defined hardware and software stack for robot bring-up, data collection, simulation, training, evaluation, and testing.
Humanoid robots often requires teams to connect separate systems for hardware integration, data capture, simulation, model training, evaluation, and deployment. The NVIDIA Isaac GR00T Reference Humanoid Robot brings these steps into one design without a closed proprietary system.
The system pairs the Unitree H2 Plus humanoid robot and Sharpa Wave tactile five-finger hands as the body. NVIDIA Jetson Thor-powered onboard compute and Isaac GR00T software workflows provide the compute and development layer for skill development and real-world validation.
“Developers want humanoid robots that are ready to build on,” said Xingxing Wang, founder and CEO of Unitree Robotics. “H2 Plus combines Unitree’s humanoid with NVIDIA Jetson Thor and the NVIDIA Isaac GR00T development platform, giving teams a validated starting point for creating robot skills and bringing them into real-world applications.”
“Humanoid robots will bring physical AI to the world’s largest industries, opening a multitrillion-dollar economic opportunity,” said Jensen Huang, founder and CEO of NVIDIA. “The NVIDIA Isaac GR00T Reference Humanoid Robot built on H2 Plus, gives researchers a single, open platform to make breakthrough discoveries toward general-purpose physical intelligence.”
A Humanoid Robot for Physical AI Development
The H2 Plus combines a human-scale robot chassis with manipulation, sensing, control, and onboard AI compute.
The reference design features:
- Unitree H2 humanoid chassis, standing nearly 6 feet tall and weighing 150 pounds, with 31 degrees of freedom across the body for human-scale testing.
- Dual Sharpa Wave tactile five-finger hands, supporting dexterous manipulation with 22 degrees of freedom and bringing the robot to 75 degrees of freedom across the body and hands.
- Multi-view sensing, including a head-mounted stereo camera with a 140-degree horizontal and 102-degree vertical field of view, wrist cameras for close-range manipulation, and an inertia measurement unit for motion tracking.
- Whole-body control, with arm torque of up to 120 Newton-meters, leg torque of up to 360 Newton-meters, a rated arm payload of 7 kilograms, and a peak payload of 15 kilograms.
- NVIDIA Jetson AGX Thor T5000 onboard compute, featuring an NVIDIA Blackwell GPU with 2,070 FP4 teraflops of AI performance, a 14-core Arm CPU, 128GB of unified memory, and a configurable 40- to 130-watt power range for real-time sensor processing and robot inference.
- Connectivity across Ethernet, Wi-Fi 6, Bluetooth 5.2, USB, microphones, and speakers for voice interaction.
- Battery for extended operation, with a 15Ah, 0.972kWh capacity and about three hours of life.
- On-remote emergency stop function for quickly disengaging the robot.
NVIDIA Isaac GR00T Software Stack for Humanoid Development
The NVIDIA software stack supports simulation, training, evaluation, and deployment while researchers retain control of robot data, training data, telemetry, and logs.
The Isaac GR00T stack includes:
- NVIDIA Isaac Teleop to capture robot demonstration data for training and policy development.
- NVIDIA Isaac GR00T open foundation models to support humanoid reasoning, learning and multitask behavior.
- NVIDIA Isaac Sim and Isaac Lab to simulate, train, test and evaluate robot policies before real-world deployment.
- Accelerated NVIDIA Isaac ROS middleware to move trained policies onto robots.
- NVIDIA Jetson Thor to run real-time, on-robot inference and control.
Its modular design lets robotics teams use the full stack or selected capabilities within existing development pipelines. This allows teams to extend humanoid development work without rebuilding the same infrastructure for each robot or task.
The NVIDIA Isaac GR00T developer stack will also support the Unitree G1 humanoid robot. That support extends the same development approach to another Unitree robot used by researchers and humanoid developers.
Availability
Unitree has released the technical specifications of the H2 Plus humanoid robot. Unitree H2 Plus will be available in late 2026.
The NVIDIA Isaac GR00T platform and reference workflow for Unitree G1 are expected to be available soon on GitHub and Hugging Face for robot developers.
Source: Unitree Robotics
About Unitree Robotics
Unitree Robotics develops and sells legged robots, humanoid robots, robotic arms, perception hardware and robot components. Founded in 2016, the company is headquartered in Hangzhou, China. Its products include Go, B, A, H, G and R series robots for consumer, education, research and industrial use. Unitree also offers six-axis manipulators, dexterous hands, 4D LiDAR, motors, servo parts, support services and software resources. The company serves universities, robotics developers, industrial inspection teams, emergency response users, entertainment operators and customers testing mobile robots. Its robots support education, research, power inspection, fire rescue, public demonstrations and humanoid development. Unitree reported shipments of more than 30,000 quadruped robots between 2022 and September 2025.
About NVIDIA
NVIDIA, founded in 1993 and headquartered in Santa Clara, CA, designs and manufactures graphics processing units, systems on chips, networking hardware, and AI intelligence software such as CUDA. Its products serve industries including gaming, data centers, autonomous vehicles, professional visualization, robotics, health care, and energy. The company introduced the GPU in 1999 and later expanded into accelerated computing and AI infrastructure. In gaming, its GPUs support high-performance rendering, while in AI and high-performance computing, its systems provide the infrastructure for training and deploying large-scale models. NVIDIA also develops tools for robotics and autonomous driving.