---
title: Wan-Alpha-I2V
canonical_url: "https://www.modelscope.cn/models/AI-ModelScope/Wan-Alpha-I2V"
md_url: "https://www.modelscope.cn/models/AI-ModelScope/Wan-Alpha-I2V.md"
repository: AI-ModelScope/Wan-Alpha-I2V
last_updated: 2026-10-20
license: apache-2.0
pipeline_tag: image-to-video
tasks:
  - image-to-video
base_model:
  - Wan-AI/Wan2.2-I2V-A14B
base_model_relation: finetune
parameters: 306.7M
tensor_type:
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 23
stars: 0
tags:
  - rgba
  - transparency
  - "effect video"
---

# Wan-Alpha-I2V

> Wan-Alpha-I2V - AI-ModelScope 在 ModelScope 开源的模型。Animating a Single RGBA Asset into Controllable Transparent Video Effects

AI-ModelScope/Wan-Alpha-I2V 是 ModelScope 魔搭社区上的 306.7M 参数image-to-video模型，采用 apache-2.0 许可，基于 Wan-AI/Wan2.2-I2V-A14B 构建。

- **Repository**: AI-ModelScope/Wan-Alpha-I2V
- **License**: apache-2.0
- **Tasks**: image-to-video
- **Parameters**: 306.7M
- **Base model**: Wan-AI/Wan2.2-I2V-A14B
- **Tags**: rgba, transparency, effect video
- **Downloads**: 23
- **Stars**: 0
- **Last updated**: 2026-10-20

Source: https://www.modelscope.cn/models/AI-ModelScope/Wan-Alpha-I2V

---

<div align="center">

  <h1>
    Wan-Alpha++
  </h1>

  <h3>Animating a Single RGBA Asset into Controllable Transparent Video Effects</h3>



[![arXiv](https://img.shields.io/badge/arXiv-2610.26189-red)](https://arxiv.org/abs/2610.26189)
[![🤗 HuggingFace](https://img.shields.io/badge/%F0%9F%A4%97%20HuggingFace-Model-orange)](https://huggingface.co/htdong/Wan-Alpha-I2V)



</div>


>Extensive experiments demonstrate that Wan-Alpha++ achieves state-of-the-art performance in image-to-RGBA generation, precise spatial transparency control, and inference efficiency.

---

### 🔥 News
* **[2026.10.19]** We release the paper of Wan-Alpha++ on [arXiv](https://arxiv.org/abs/2610.26189).
* **[2026.10.19]** Released checkpoints for normal RGBA video of Wan-Alpha++ on [GitHub](https://github.com/WeChatCV/Wan-Alpha_PlusPlus).


---

### 📝 To-Do List

- [x] **Paper**: Available on [arXiv](https://arxiv.org/abs/2610.26189).
- [x] **Inference Code**: Released checkpoints for normal RGBA video of Wan-Alpha++ on [GitHub](https://github.com/WeChatCV/Wan-Alpha_PlusPlus).
- [x] **Model Weights of Normal RGBA Video**: Released checkpoints for Wan-Alpha++.
- [] Training Code.
- [] Checkpoints for Effect RGBA Video.




### 🚀 Quick Start

##### 1. Environment Setup
```bash
# Clone the project repository
git clone https://github.com/WeChatCV/Wan-Alpha_PlusPlus.git
cd Wan-Alpha++

# Create and activate Conda environment
conda create -n Wan-Alpha++ python=3.11 -y
conda activate Wan-Alpha++

# Install dependencies
pip install -r requirements.txt
```

##### 2. Model Download
Download [Wan2.2-I2V-A14B](https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B)


Download [Wan-Alpha++](https://huggingface.co/htdong/Wan-Alpha-I2V)

Download [Wan-Alpha-v2.0-VAE](https://huggingface.co/htdong/Wan-Alpha-VAE/blob/main/Wan-Alpha-v2-VAE.bin)

### 🧪 Usage
You can test our model through:
```bash
torchrun --nproc_per_node=8  generate_2.2_new.py \
        --task i2v-A14B --size 832*480 \
        --ckpt_dir Wan-AI/Wan2.2-I2V-A14B \
        --dit_fsdp --t5_fsdp --ulysses_size 8 \
        --frame_num 49 \
        --vae_lora_checkpoint "VAE/pytorch_model.bin" \
        --high_lora high_noise.safetensors \
        --low_lora low_noise.safetensors \
        --output_dir $output_path/result
```
You can specify the weights of `Wan2.2-I2V-A14B` with `--ckpt_dir`, `Wan-Alpha-VAE` with `--vae_lora_checkpoint`, and `Wan-Alpha-I2V` with `--high_lora` and `--low_lora`. Finally, you can find the rendered RGBA videos with a checkerboard background and PNG frames at `--output_dir`.


**Prompt Writing Tip:**  You need to specify that the background of the video is transparent, the visual style, the shot type (such as close-up, medium shot, wide shot, or extreme close-up), and a description of the main subject. Prompts support both Chinese and English input.

```bash
# An example of prompt.
This video has a transparent background. A woman holds a clear cup containing orange juice; she lifts the cup, takes a sip, and smiles.
```




### 🤝 Acknowledgements

This project is built upon the following excellent open-source projects:
* [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio) (training/inference framework)
* [Wan2.2](https://github.com/Wan-Video/Wan2.2) (base video generation model)


We sincerely thank the authors and contributors of these projects.


### ✏ Citation

If you find our work helpful for your research, please consider citing our paper:

```bibtex
@misc{dong2026wanalphaanimatingsinglergba,
      title={Wan-Alpha++: Animating a Single RGBA Asset into Controllable Transparent Video Effects}, 
      author={Haotian Dong and Xin Wang and Wenjing Wang and Chen Li and Jing Lyu and Yan Tina Luximon and Ping Li and Wanchao Su and Ke Xu and Di Lin},
      year={2026},
      eprint={2610.26189},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2610.26189}, 
}
``` 

### 📬 Contact Us

If you have any questions or suggestions, feel free to reach out via [GitHub Issues](https://github.com/WeChatCV/Wan-Alpha_PlusPlus/issues) . We look forward to your feedback!