Crilias/Kaleido-14B-S2V

KALEIDO: OPEN-SOURCED MULTI-SUBJECT REFERENCE VIDEO GENERATION MODEL

This repository contains the official implementation of Kaleido, proposed in our paper:

Update and News

• 2025.10.28: 🔥 We release the checkpoints of Kaleido-14B-S2V.
• 2025.10.22: 🔥 We propose Kaleido, a novel multi-subject reference video generation model.

Checkpoints Download

ckpts	Download Link	Notes
Kaleido-14B	🤗 Huggingface	Supports 512P

Use the following commands to download the model weights (We have integrated both Wan VAE and T5 modules into this checkpoint for convenience).

# Download the repository (skip automatic LFS file downloads)
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/Crilias/Kaleido-14B-S2V

# Enter the repository folder
cd Kaleido-14B-S2V

# Merge the checkpoint files
python merge_kaleido.py

Arrange the model files into the following structure:

.
├── Kaleido-14B-S2V
│   ├── model
│   │   └── ....
│   ├── Wan2.1_VAE.pth
│   │
│   └── umt5-xxl
│       └── ....
├── configs
├── sat
└── sgm

Citation

If you find our work helpful, please cite our paper:

@misc{zhang2025kaleidoopensourcedmultisubjectreference,
      title={Kaleido: Open-Sourced Multi-Subject Reference Video Generation Model}, 
      author={Zhenxing Zhang and Jiayan Teng and Zhuoyi Yang and Tiankun Cao and Cheng Wang and Xiaotao Gu and Jie Tang and Dan Guo and Meng Wang},
      year={2025},
      eprint={2510.18573},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2510.18573}, 
}