Unitree H1

Open-world humanoid manipulation across dexterous object manipulation, human-humanoid interaction and loco-manipulation. The dataset is a training resource; the planned cloud evaluation would score deployed policies on real humanoids.

Listed benchmark embodiment · Collected on two Unitree humanoids: the G1 (29 DoF, with 7-DoF three-finger Dex3-1 hands with tactile sensing) and the H1 (27 DoF, with 6-DoF Inspire hands). Both carry Intel RealSense RGB-D cameras and a Livox LiDAR, according to the paper.

Whether a learning method can control a high-dimensional humanoid to walk, run, stand, balance and climb stairs, and to carry out whole-body manipulation such as pushing, opening doors and cabinets, and moving packages. Performance is reported as episode return per task.

Listed benchmark embodiment · Primary robot: simulated Unitree H1 with two Shadow Dexterous Hands on the arms. The repository also supports the Unitree G1 with three-finger hands and a low-dimensional-hand H1 variant. The paper also lists the Agility Digit as a supported body; it is not linked here.

How well imitation-learning methods reproduce agile locomotion from motion-capture references, scored with handcrafted, task-specific metrics. The repository's documentation names trajectory comparison metrics such as dynamic time warping and Fréchet distance.

Listed benchmark embodiment · The README names Unitree H1 ('UnitreeH1') and Unitree G1 ('MjxUnitreeG1') among its humanoid environments. Quadruped and biomechanical models (MyoSkeleton) are also included. Other humanoids, such as Atlas and Talos, are not named in the sources opened.

Generalizable manipulation and locomotion skills in a fast, GPU-parallel simulator, with task success measured per environment. The paper emphasizes simulation throughput alongside task performance.

Listed benchmark embodiment · The ManiSkill documentation lists Unitree G1 and H1 models (including simplified leg and upper-body variants) among supported robots. Most listed robots are arms and mobile manipulators. Which tasks use the humanoid models was not confirmed.

Task accuracy and motion feasibility for whole-body loco-manipulation, such as reaching, pressing buttons, pushing boxes, carrying packages and shooting a ball, across humanoid bodies. The paper also argues that its method avoids reward hacking.

Listed benchmark embodiment · Three Unitree humanoids: H1, G1 and H1-2, mapped to the roster slugs unitree-h1, unitree-g1 and unitree-h1-2. Degrees of freedom for each embodiment are not given on the project page or in the abstract.

Sprint and endurance locomotion, balance, team football, dexterous manipulation, and performance in household, industrial and emergency-response scenarios. Control-mode and autonomy requirements differ by event and edition.

Listed benchmark embodiment · Tiangong Ultra won the 2025 100 m, Unitree H1 won the 2025 400 m and 1500 m, and Booster T1 teams took the 2025 5v5 football gold and silver medals in the collected results.