TSXu/UniCalli-base

Official pre-trained model for UniCalli: A Unified Diffusion Framework for Column-Level Generation and Recognition of Chinese Calligraphy

Model Details

• Base Architecture: FLUX
• Model Size: ~23GB
• Training Data: Large-scale Chinese calligraphy dataset with column-level annotations
• Supported Tasks:
  • Column-level calligraphy generation
  • Calligraphy text recognition (OCR)

License

For academic research and non-commercial use only. For commercial use, please contact the authors.

本模型仅供学术研究、非商业使用，商业使用请联系作者。

Key Features

• 🎨 Unified Architecture: First framework to unify column-level calligraphy generation and recognition
• ✍️ Multi-Master Styles: Supports diverse calligraphic styles, including Wang Xizhi (王羲之), Yan Zhenqing (颜真卿), Ouyang Xun (欧阳询), etc.
• 📚 Densely Annotated Data: Trained on large-scale calligraphy dataset with detailed annotations.

Usage

Download Model

from huggingface_hub import hf_hub_download

# Download the model weights
model_path = hf_hub_download(
    repo_id="TSXu/UniCalli-base",
    filename="unicalli-base_cleaned.bin"
)

Prerequisites

The full inference code and usage examples will be released soon. Please check the GitHub repository for updates.

Citation

If you find UniCalli useful in your research, please consider citing:

@article{xu2025unicalli,
  title={UniCalli: A Unified Diffusion Framework for Column-Level Generation and Recognition of Chinese Calligraphy},
  author={Xu, Tianshuo and Wang, Kai and Chen, Zhifei and Wu, Leyi and Wen, Tianshui and Chao, Fei and Chen, Ying-Cong},
  journal={arXiv preprint arXiv:2025.13745},
  year={2025}
}

Acknowledgments

This work builds upon the FLUX architecture and benefits from the rich heritage of Chinese calligraphy. We thank the calligraphy masters whose works made this research possible.