zai-org/SCAIL-Preview

SCAIL: Towards Studio-Grade Character Animation via In-Context Learning of 3D-Consistent Pose Representations

This repository contains the model weights for SCAIL (Studio-Grade Character Animation via In-Context Learning), a framework that enables high-fidelity character animation under diverse and challenging conditions, including large motion variations, stylized characters, and multi-character interactions.

🔎 Project Page

Check our model architecture design, our video demo, as well as more comparisons against other baselines at this link, more creative examples will be added to the gallery soon.

📋 TODOs

• Model Weights for Preview 14B SCAIL Model(512p)
• Model Weights for Official 1.3B/14B SCAIL Model(720p with history support)

📋 Note

This repository only contains the model weights for the SCAIL model, for model inference, please refer to the official repository. For pose extraction, please refer to the SCAIL-Pose.

📄 Citation

If you find this work useful in your research, please cite:

@article{yan2025scail,
  title={SCAIL: Towards Studio-Grade Character Animation via In-Context Learning of 3D-Consistent Pose Representations},
  author={Yan, Wenhao and Ye, Sheng and Yang, Zhuoyi and Teng, Jiayan and Dong, ZhenHui and Wen, Kairui and Gu, Xiaotao and Liu, Yong-Jin and Tang, Jie},
  journal={arXiv preprint arXiv:2512.05905},
  year={2025}
}