Skip to content

README.md

Latest commit

 

History

History
64 lines (43 loc) · 2.51 KB

README.md

File metadata and controls

64 lines (43 loc) · 2.51 KB
Raw

ComfyUI Load Images Browse Folder

A ComfyUI custom node that allows you to browse and select a folder, then load and process all images from that folder one at a time through the complete workflow (like VideoHelperSuite's meta-batch processing).

Features

  • Folder browsing interface - Select folders through a user-friendly UI
  • Automatic meta-batch processing - Each image runs through the entire workflow independently
  • One image at a time - Perfect for preventing memory issues and processing images sequentially
  • Automatic sorting - Optional alphabetical sorting of images
  • Format support - Supports JPG, PNG, BMP, WebP, and TIFF image formats

Installation

  1. Clone this repository into your ComfyUI custom_nodes directory:
cd ComfyUI/custom_nodes
git clone <repository-url> comfyui-load-images-browse-folder
  1. Restart ComfyUI

Usage

Simply add the "Load Images From Folder (Browse)" node to your workflow:

  1. Enter a folder path (or use the browse functionality if available)
  2. The node will automatically process one image at a time through your entire workflow
  3. If you have 10 images, the workflow will run 10 times (once per image)

How it works differently from regular batch loading:

Traditional batch loading:

  • All images load at once → All process through step 1 → All process through step 2 → etc.
  • Preview nodes show all images in a grid

This node (meta-batch):

  • Image 1: Load → Process completely through workflow → Save
  • Image 2: Load → Process completely through workflow → Save
  • Image 3: Load → Process completely through workflow → Save
  • ...and so on

This is perfect for workflows where you want each image to be processed independently, matching VideoHelperSuite's behavior.

Node Parameters

Load Images From Folder (Browse)

  • folder_path (required): Path to the folder containing images
  • sort (optional, default: True): Sort images alphabetically

Outputs:

  • IMAGE: Currently loaded image (1 image per workflow execution)
  • frame_count: Number of images in this batch (always 1)

Example Use Cases

  • Sequential img2img processing: Process each source image through the pipeline independently
  • Batch upscaling: Upscale images one at a time to avoid memory issues
  • Iterative processing: Any workflow where each image needs independent processing

Credits

Meta-batch processing implementation inspired by ComfyUI-VideoHelperSuite.