ComfyUI-Direct3D-S2

ComfyUI-Direct3D-S2
★ 17

3D生成稀疏体积空间稀疏注意力降低训练成本
在ComfyUI中集成Direct3D‑S2，基于稀疏体积和空间稀疏注意力实现可扩展的千兆级3D生成，提升输出质量并显著降低训练成本。
💡 在ComfyUI中进行高质量、低成本的大规模3D模型生成。
🍴 5 Forks💻 Python🔄 2025-06-10
📦
网盘下载
复制链接后前往夸克网盘下载
https://pan.quark.cn/s/a9fb3a59e10c
📦 requirements.txt
scikit-image
trimesh
omegaconf
tqdm
huggingface_hub
einops
numpy
transformers==4.40.2
diffusers
triton==3.1.0
flash-attn
pymeshfix
utils3d
pyvista
igraph
third_party/voxelize/
📄 README

ComfyUI-Direct3D-S2

ComfyUI-Direct3D‑S2 is now available in ComfyUI, Direct3D‑S2 – Gigascale 3D Generation Made Easy with Spatial Sparse Attention. Direct3D‑S2 is a scalable 3D generation framework based on sparse volumes that achieves superior output quality with dramatically reduced training costs.

Installation

  • Make sure you have ComfyUI installed

  • Clone this repository into your ComfyUI’s custom_nodes directory:
    • cd ComfyUI/custom_nodes
git clone https://github.com/Yuan-ManX/ComfyUI-Direct3D-S2.git

  • Install dependencies:
    • cd ComfyUI-Direct3D-S2

# Make sure the PyTorch CUDA version matches your installed CUDA Toolkit.
pip install torch==2.5.1 torchvision==0.20.1 --index-url https://download.pytorch.org/whl/cu121

# Torchsparse
git clone https://github.com/mit-han-lab/torchsparse
cd torchsparse && python -m pip install .

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

    Model

    Download Pretrained Models

    Direct3D-S2 Pretrained Models