Xiaomi-Robotics-1

Scaling Vision-Language-Action Models with over 100K Hours of Real-World Trajectories

Our latest robot foundation model, pre-trained on over 100K hours of real-world trajectories, brings scaling to embodied intelligence—and is ready to use out of the box.

🚀 Coming soon — July 16, 2026

Code and model weights will be released soon. Stay tuned.

Citation

If you find this work useful, please consider citing:

@article{xiaomi2026robotics1,
  title={Xiaomi-Robotics-1: Scaling Vision-Language-Action Models with over 100K Hours of Real-World Trajectories},
  author={Xiaomi Robotics},
  journal={arXiv preprint},
  year={2026}
}