1. 打开Terminal

  2. 安装Xcode

    xcode-select --install

xcode-select: error: command line tools are already installed, use "Software Update" in System Settings to install updates
出现这行，说明已经安装完成了

  3. 对版本

    python3 --version
Python 3.11.5

pip3 --version
pip 23.2.1 from /opt/homebrew/lib/python3.11/site-packages/pip (python 3.11)

git --version
git version 2.39.2 (Apple Git-143)

brew install python3
brew install git

  4. 克隆Repo

    <https://github.com/comfyanonymous/ComfyUI>

  5. 安装虚拟环境

    cd ComfyUI
python3 -m venv venv
然后激活
source venv/bin/activate

  6. 安装Pytorch Nightly Build

    # 注意，一定要安装nightly build，不能安装官网上的版本。
pip3 install --pre torch torchvision torchaudio --extra-index-url <https://download.pytorch.org/whl/nightly/cpu>
pip install git+https://github.com/pytorch/pytorch@refs/pull/114183/head git+https://github.com/pytorch/vision git+https://github.com/pytorch/audio

  7. 下载模型

    # SDXL Turbo
<https://huggingface.co/stabilityai/sdxl-turbo/tree/main>

# Stable Video Diffusion
14 framerate
<https://huggingface.co/stabilityai/stable-video-diffusion-img2vid/tree/main>
25 framerate
<https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt/tree/main>

下载完成后把模型复制到这个目录里
cd models/checkpoints

  8. 安装comfiyui manager

    cd custom_nodes
git clone <https://github.com/ltdrdata/ComfyUI-Manager>

  9. 安装Python 依赖包

    cd ..
pip install -r requirements.txt

  10. 加载模型

    cd models/checkpoints

  11. 启动ComfyUI

    python main.py --force-fp16

  12. 加载 Work Flow

  13. 点击Manager，然后安装custom nodes

  14. 加载模型，就可以使用了