🌌 DreamX-Phi

DreamX-Phi 1.0: Action-Conditioned Video World Model for Robotic Manipulation

DreamX-Phi is a geometry-aware video world model built on Wan2.2-TI2V-5B. Given an initial observation and a prescribed robot action sequence, it predicts the corresponding future video while preserving robot motion, scene geometry, and the state of the manipulated object.

✨ Highlights

• Geometry-aware action conditioning for end-effector poses and gripper states.
• Interaction-aware learning for coherent scene and object evolution.
• DMD-based few-step generation for efficient inference.
• First place on WorldArena 2.0 Track 1 and tied for second on Track 2.

🚀 Release

Model weights and inference code will be made publicly available after the WorldArena 2.0 IROS Challenge concludes.

📖 Citation

Read the technical report on arXiv. If you find this work useful, please cite:

@article{team2026dreamx,
  title   = {DreamX-Phi 1.0: Action-Conditioned Video World Model for Robotic Manipulation},
  author  = {Team, DreamX and Chen, Rui and Chu, Xiangxiang and Li, Geng and Li, Jifan and Shi, Qingfeng and Tang, Datao and Tang, Jing and Wang, Jun and Zhang, Pengfei},
  journal = {arXiv preprint arXiv:2608.13489},
  year    = {2026}
}

📄 License

This repository is released under the MIT License.