Seb-Lis c111b84f2a Revise README for Custom Nodes for ComfyUI
Updated the README to reflect the new project name and added installation instructions, node descriptions, and example workflow.
2025-11-30 12:08:30 +01:00
2025-11-30 12:08:10 +01:00

Custom Nodes for ComfyUI

A collection of utility nodes for image processing, workflow management, and generation metadata tracking.

Installation

  1. Navigate to your ComfyUI custom_nodes directory:
    cd ComfyUI/custom_nodes/
    
  2. Clone this repository (or copy the folder seb-nodes-comfyui):
    git clone https://github.com/yourusername/seb-nodes-comfyui.git
    
  3. Restart ComfyUI.
    • Connect the outputs of this node (Seed, Steps, etc.) to your actual KSampler node.
    • Connect text_overlay and start_time to the Generation Time node (optional).

2. Generation Time (SEB)

Category: sampling/control

Calculates the total generation time and appends it to your text overlay.

  • Inputs:
    • image: The generated image (from VAE Decode).
    • text: The text_overlay output from the KSampler Bridge node.
    • start_time: The start_time output from the KSampler Bridge node.
  • Outputs:
    • image: Passthrough of the input image.
    • text: Updated text string including "Time: X.XXs".
  • How to Connect: Place this after your VAE Decode. Connect the image and the data from the Bridge node.

3. Text Label (SEB)

Combines multiple image batches into a single grid layout.

  • Inputs: columns (grid width), padding, and up to 6 optional image inputs (image_1...image_6).
  • Outputs: A single grid image.
  • Usage: Useful for comparing results or creating contact sheets.

Example Workflow

  1. KSampler Bridge -> (Pass parameters) -> KSampler -> VAE Decode -> Generation Time (Input Image).
  2. KSampler Bridge -> (Pass text_overlay + start_time) -> Generation Time.
  3. Generation Time -> (Pass image + text) -> Text Label.
  4. Text Label -> Save Image.
S
Description
No description provided
Readme MIT
2 MiB
Languages
Python 100%