OmniXtreme
Preprint

OmniXtreme Breaking the Generality Barrier in High-Dynamic Humanoid Control

Yunshen Wang1,2,3,* Shaohang Zhu1,2,4,* Peiyuan Zhi1,2 Yuhan Li1,2,6 Jiaxin Li1,2,7
Yong-Lu Li3 Yuchen Xiao5 Xingxing Wang5 Baoxiong Jia1,2,† Siyuan Huang1,2,†

1 Beijing Institute for General Artificial Intelligence (BIGAI)
2 Joint Laboratory of Embodied AI and Humanoid Robots, BIGAI & UniTree Robotics
3 Shanghai Jiao Tong University, 4 University of Science and Technology of China, 5 UniTree Robotics
6 Huazhong University of Science and Technology, 7 Beijing Institute of Technology
* Equal contribution. † Corresponding authors.

Paper (Incoming) ArXiv (Incoming) Code (Incoming) Video (Incoming)

Behavior Montage

Behavior Montage

Method

Method figure
Pre-training

A unified base policy is trained via DAgger-based Flow Matching to aggregate diverse motion priors from different motion tracking experts.

Post-training

The base policy is frozen while a residual policy is optimized under stringent motor constraints, extensive domain randomization, and power-safety regularization to bridge the sim-to-real gap.

Deployment

The whole inference pipeline is real- time and executed entirely onboard, facilitating robust and agile control in physical environments.

Ablation Study

UA ablation figure

Power-Safety Regularization explicitly penalizes excessive negative joint power to prevent unsafe energy absorption and transient braking loads during high-dynamic motions.

Citation

Use this project-page BibTeX entry for reference.

@misc{omnixtreme_project_2026,
  title={OmniXtreme Project Page},
  author={Yunshen Wang and Shaohang Zhu and Peiyuan Zhi and Yuhan Li and Jiaxin Li and Yong-Lu Li and Yuchen Xiao and Xingxing Wang and Baoxiong Jia and Siyuan Huang},
  year={2026},
  howpublished={\url{https://extreme-humanoid.github.io}},
  note={Accessed: 2026-02-24}
}