ComfyUI InfoComfyUI Info

ComfyUI-Impact-Subpack

The ComfyUI-Impact-Subpack Node Package is an extension for the ComfyUI Impact Pack, featuring the UltralyticsDetectorProvider. It has core nodes like UltralyticsDetectorProvider to generate detector outputs, manages models by type in specific directories sourced from multiple channels, allows path customization and has security protocols, can be installed via ComfyUI-Manager or manually, and credits ComfyUI and various model resources. Its main functions include enhancing ComfyUI Impact Pack functionality, providing detector outputs, managing models, and ensuring secure model loading with multiple installation options.

ltdrdata

v1.3.2
203 stars
GitHub

Description

ComfyUI-Impact-Subpack

This node pack provides nodes that complement the ComfyUI Impact Pack, such as the UltralyticsDetectorProvider.

Nodes

  • UltralyticsDetectorProvider - Loads the Ultralystics model to provide SEGM_DETECTOR, BBOX_DETECTOR.
  • Unlike MMDetDetectorProvider, for segm models, BBOX_DETECTOR is also provided.
  • The various models available in UltralyticsDetectorProvider can be downloaded through ComfyUI-Manager.

Ultralytics models

  • When using ultralytics models, save them separately in models/ultralytics/bbox and models/ultralytics/segm depending on the type of model. Many models can be downloaded by searching for ultralytics in the Model Manager of ComfyUI-Manager.
  • huggingface.co/Bingsu/adetailer - You can download face, people detection models, and clothing detection models.
  • ultralytics/assets - You can download various types of detection models other than faces or people.
  • civitai/adetailer - You can download various types detection models....Many models are associated with NSFW content.

Paths

  • In extra_model_paths.yaml, you can add the following entries:
  • ultralytics_bbox - Specifies the paths for bbox YOLO models.
  • ultralytics_segm - Specifies the paths for segm YOLO models.
  • ultralytics - Allows the presence of bbox/ and segm/ subdirectories.
  • Loading model files can involve executing code, so if malicious code is embedded in a model, it can pose a security risk. For this reason, PyTorch 2.6 and later have introduced features for safer model loading. The issue is that older model files, which were created without such safe loading (weights_only) restrictions, may occasionally fail to load properly.
  • By listing the paths of models deemed safe in <user_directory>/default/ComfyUI-Impact-Subpack/model-whitelist.txt, those specific models will have weights_only disabled, allowing them to be loaded without restriction.
    • <user_directory>: typically located at ComfyUI/user.

How To Install?

  • Search ComfyUI Impact Subpack in ComfyUI-Manager and click Install button.
  1. cd custom_nodes
  2. git clone https://github.com/ltdrdata/ComfyUI-Impact-Subpack
  3. cd ComfyUI-Impact-Subpack
  4. pip install -r requirements.txt
    • IMPORTANT:
      • You must install it within the Python environment where ComfyUI is running.
      • For the portable version, use <installed path>\python_embeded\python.exe -m pip instead of pip. For a venv, activate the venv first and then use pip.
  5. Restart ComfyUI

Credits

ComfyUI/ComfyUI - A powerful and modular stable diffusion GUI.

Bing-su/adetailer - This repository provides an object detection model and features based on Ultralystics.

huggingface/Bingsu/adetailer - This repository offers various models based on Ultralystics.

  • You can download other models supported by the UltralyticsDetectorProvider from here.