ROBOTERA Launches XHAND 1 Pro-U, Featured in NVIDIA’s GR00T 2 DexUMI Demonstration
![]() |
BEIJING, Nov. 16, 2026 /PRNewswire/ — ROBOTERA today announced the commercial launch of XHAND 1 Pro-U, a high-degree-of-freedom five-finger dexterous hand featured in a GR00T 2 demonstration presented by Jim Fan, Director and Distinguished Research Scientist at NVIDIA, at the Conference on Robot Learning (CoRL).
Designed for dexterous manipulation, XHAND 1 Pro-U features a human-like configuration and a 16-degree-of-freedom (DoF) mechanical structure closely aligned with NVIDIA GEAR Lab’s DexUMI exoskeleton. Its design has been refined to support both natural human demonstrations and reliable robot execution, with corresponding joints that allow hand movements to map directly between the exoskeleton and the robotic hand. These refinements reduce the need for additional motion conversion and help preserve the details of demonstrated movements. The redesign also optimizes thumb positioning, palm-side tactile sensing, and hand articulation to better accommodate natural demonstrations and varied grasping postures.
Together, DexUMI and XHAND 1 Pro-U enable a complete demonstration-to-deployment workflow, connecting human skill capture, model post-training, and real-world robot execution. Wearing the exoskeleton, users demonstrate tasks by directly manipulating real objects. DexUMI records hand movements alongside fingertip force information, capturing not only how an action is performed but also how force is applied during physical contact. These demonstrations are used to post-train GR00T 2, and the resulting skills are transferred to a dual-arm robot equipped with two XHAND 1 Pro-U hands. Unlike conventional teleoperation, this approach does not require an operator to remotely control the target robot during data collection. For the tasks shown, post-training used DexUMI demonstrations without additional task-specific teleoperation or extra robot play data.
The system was evaluated on three multi-step tasks: preparing banana yogurt, scooping granola, and fire chemistry, which involved striking a match, lighting a lamp, and placing the match in water. GR00T 2 achieved overall success rates of 60%, 80%, and 75%, respectively. These tasks involved a range of dexterous skills, including tool use, object transfer, lid handling, and coordinated two-handed actions, demonstrating the system’s ability to execute complex, multi-step tasks through physical interaction.
XHAND 1 Pro-U is now available for commercial orders, with an initial limited release of 40 units. ROBOTERA welcomes inquiries from research teams, robotics developers, and industry partners working on dexterous manipulation, robot learning, and real-world robotic applications.



