/ComfyUI-Allor

ComfyUI plugin for image processing and work with alpha chanel.

Primary LanguagePythonMIT LicenseMIT

logo

This image from start to end was done in ComfyUI. How?
  1. Install plugin.
  2. Load fonts Overlock SC and Merienda.
  3. Put OverlockSC-Regular.ttf and Merienda-Regular.ttf in to fonts folder.
  4. Load RealESRNet_x4plus.pth and put in to models/upscale folder.
  5. Load AOM3A1B_orangemixs.safetensors and put in to models/chekpoints folder.
  6. Load orangemix.vae.pt and put in to models/vae folder.
  7. Drag-and-drop this image to ComfyUI or load JSON.
  8. Press the Queue Promt button.

About Allor:

  • Allor is a high-performance ComfyUI plugin designed for image processing.
  • It comprises more than 90 nodes, each with numerous parameters for your needs.
  • It supports transparency and multi-image processing across all modules and nodes.
  • Allor is fully configurable, offering the option to disable any functionality that is not required.
  • The majority of its operations are implemented in tensor space to avoid unnecessary data transformations.

Documentation

We are pleased to present to you our new documentation.

The documentation includes information about installation, updating, all nodes including their parameters and much more.

Contributing:

ComfyUI Allor is an open source project and I welcome any contributions from the community.

  • Suggest ideas: If you have thoughts on improving this plugin, feel free to share them on GitHub.

  • Report bugs: If you find a bug or problem, please report it in the Issue with steps to reproduce it.

  • Make pull requests: If you’d like to help with the plugin’s code, you can copy this repository, make your changes, and I’ll check and include them if they work well with the plugin.

Jupiter Notebook:

To run Allor on another cloud service, you can use this Jupiter Notebook.

If you have a subscription to Colab Pro, then you can run Allor in Google Colab.