README.md

May 26, 2025 · View on GitHub

Wan Toy Transform



Alibaba Research Intelligence Computing

This is a LoRA model finetuned on Wan-I2V-14B-480P. It turns things in the image into fluffy toys. 🌟 Give it a star if you like it.

🎞️ Showcases

🐍 Installation

# Python 3.12 and PyTorch 2.6.0 are tested.
pip install torch==2.6.0 torchvision==0.21.0 --index-url https://download.pytorch.org/whl/cu124
pip install -r requirements.txt

🔄 Inference

python generate.py --prompt "The video opens with a clear view of a $name. Then it transforms to a b6e9636 JellyCat-style $name. It has a face and a cute, fluffy and playful appearance." --image $image_path --save_file "output.mp4" --offload_type leaf_level

Note:

  • Change $name to the object name you want to transform.

  • $image_path is the path to the first frame image.

  • Choose --offload_type from ['leaf_level', 'block_level', 'none', 'model']. More details can be found here.

  • VRAM usage and generation time of different --offload_type are listed below.

    --offload_typeVRAM UsageGeneration Time (NVIDIA A100)
    leaf_level11.9 GB17m17s
    block_level (num_blocks_per_group=1)20.5 GB16m48s
    model39.4 GB16m24s
    none55.9 GB16m08s

🤝 Acknowledgements

Special thanks to these projects for their contributions to the community!

📄 Our previous work