Skip to content

Latest commit

 

History

History
24 lines (18 loc) · 1.76 KB

manual-setup.adoc

File metadata and controls

24 lines (18 loc) · 1.76 KB
Raw

How To Manually Setup ComfyUI

If you want to setup ComfyUI (with 3D-Pack) from the ground up, the key steps are:

  1. Setup OS environment: Git, CUDA Toolkit, C++ build tools.

  2. Setup Python venv, or embeded. Then setup PIP, or conda.

  3. Install PyTorch + xFormers. Each version of xFormers is paired with specific PyTorch version, so you could choose the PyTorch version by choosing xFormers' version.

  4. Git clone ComfyUI and custom nodes.

  5. Install requirements.txt from ComfyUI. When done, a basic environment for running SD image generation is ready.

  6. Install requirements.txt from custom nodes. Which is overwhelmingly full of details. And you should look into them one by one.

    • numpy needs 1.x version.

    • onnxruntime-gpu needs a specific repo to download wheels from.

    • cupy-cuda12x needs manual installation.

    • ninja needs binary from its official, or just OS install and configure the PATH.

    • dlib and insightface needs to compile on site, or download binary wheel files from 3rd party.

    • CLIP needs to install from source.

    • gradio is incompatible with ComfyUI-Manager as they depends on different versions of urllib3.

  7. Comfy3D’s dependency is complicated and needs special attention. You can use its installer, but manual installation gives you more control. You should examine the scripts from top to bottom, pick the proper wheels to install, and copy the includes and libs for later C++ compiling.

  8. Finally, put models in and let Python run the main.py.

For details, check the source code of this repository.