Consistent and Controllable Image Animation with Motion Linear Diffusion Transformers
Xin Ma, Yaohui Wang*, Gengyun Jia, Xinyuan Chen, Tien-Tsin Wong, Cunjian Chen*
(*Corresponding authors)

This repo contains pre-trained weights, and sampling code of MiraMo. Please visit our project page for more results.

• 🔥 Jul. 31, 2025 💥 The inference code is released. The pre-trained models will come soon. Stay tuned!

Download and set up the repo:

git clone https://github.com/maxin-cn/MiraMo
cd MiraMo
conda env create -f environment.yml
conda activate miramo

Launching soon. Please stay tuned!

• T2V Inference
• T2V Checkpoints
• I2V Inference
• I2V Checkpoints
• Training

If you find this work useful for your research, please consider citing it.

@article{ma2025consistent,
  title={Consistent and Controllable Image Animation with Motion Linear Diffusion Transformers},
  author={Ma, Xin and Wang, Yaohui and Jia, Genyun and Chen, Xinyuan and Wong, Tien-Tsin and Chen, Cunjian},
  journal={arXiv preprint arXiv:2508.07246},
  year={2025}
}

MiraMo has been greatly inspired by the following amazing works and teams: Latte and SANA. We thank all the contributors for open-sourcing.

The code and model weights are licensed under LICENSE.