ComfyUI Extension: Flux Prompt Generator for ComfyUI

Authored by fairy-root

Created

Updated

142 stars

A flexible and customizable prompt generator for generating detailed and creative prompts for image generation models for ComfyUI

Custom Nodes (0)

    README

    Flux-Prompt-Generator

    Flux Prompt Generator is a ComfyUI node that provides a flexible and customizable prompt generator for generating detailed and creative prompts for image generation models. based on the work by Aitrepreneur found here Flux Prompt Generator python code and modified to work with ComfyUI by FairyRoot

    Flux Prompt Generator

    Overview

    The Flux Prompt Generator utilizes a collection of JSON data files containing various categories of descriptive terms. These categories include:

    • Artform: Photography, Digital Art, etc.
    • Photo Type: Portrait, Landscape, etc.
    • Body Types: Muscular, Thin, etc.
    • Default Tags: Man, Woman, Child, etc.
    • Roles: Knight, Wizard, etc.
    • Hairstyles: Long, Short, Braided, etc.
    • Additional Details: Clothing, Accessories, etc.
    • Photography Styles: Cinematic, Realistic, etc.
    • Device: Camera models, etc.
    • Photographer: Famous photographers, etc.
    • Artist: Famous artists, etc.
    • Digital Artform: Pixel Art, 3D Render, etc.
    • Place: Forest, City, etc.
    • Lighting: Soft, Harsh, etc.
    • Clothing: Dress, Suit, etc.
    • Composition: Rule of Thirds, Golden Ratio, etc.
    • Pose: Standing, Sitting, etc.
    • Background: Plain, Detailed, etc.
    • Face Features: Sharp Jawline, High Cheekbones, etc.
    • Eye Colors: Blue, Green, Brown, etc.
    • Facial Hair: Beard, Mustache, etc.
    • Skin Tone: Pale, Dark, etc.
    • Age Group: Child, Teenager, Adult, etc.
    • Ethnicity: Various ethnicities and cultural backgrounds.
    • Accessories: Glasses, Hats, Jewelry, etc.
    • Expression: Smile, Frown, Surprise, etc.
    • Tattoos & Scars: Detailed descriptions of body modifications.
    • Makeup Styles: Natural, Glamorous, etc.
    • Hair Color: Blonde, Brown, Black, etc.
    • Body Markings: Birthmarks, Moles, Freckles, etc.

    Flux Prompt Generator

    The node allows you to select specific terms or choose "random" to let the generator pick random terms from the corresponding JSON file. This randomness adds a degree of unpredictability and creativity to the generated prompts.

    Installation

    1. cd to the custom_nodes folder inside of ComfyUI directory
    2. cmd in the address bar, then use this command:
    git clone https://github.com/fairy-root/Flux-Prompt-Generator.git
    

    Usage

    1. Add the "Flux Prompt Generator" node to your ComfyUI workflow.

    2. Configure the desired parameters:

      • Seed: Controls the randomness of the generator.
      • Custom: Add any custom text to the prompt.
      • Subject: Specify the main subject of the image.
      • Artform: Choose the desired art form (Photography, Digital Art, etc.).
      • ... (All other parameters as described in the Overview section)
    3. Connect the output of the node to a text-to-image model (like Flux or Stable Diffusion...etc) to generate images based on the generated prompt.

    Example

    Let's say you want to generate a prompt for a portrait photograph of a woman with long hair, wearing a dress, and standing in a forest. You could configure the node with the following parameters:

    • Artform: Photography
    • Photo Type: Portrait
    • Default Tags: Woman
    • Hairstyles: Long Hair
    • Clothing: Dress
    • Place: Forest

    The node would then generate a prompt similar to: "photography of a woman with long hair, dressed in a dress, in a forest."

    Donation

    Your support is appreciated:

    • USDt (TRC20): TGCVbSSJbwL5nyXqMuKY839LJ5q5ygn2uS
    • BTC: 13GS1ixn2uQAmFQkte6qA5p1MQtMXre6MT
    • ETH (ERC20): 0xdbc7a7dafbb333773a5866ccf7a74da15ee654cc
    • LTC: Ldb6SDxUMEdYQQfRhSA3zi4dCUtfUdsPou

    Author and Contact

    License

    This project is licensed under the MIT License. See the LICENSE file for details.

    Contributing

    Contributions are welcome! Please open an issue or submit a pull request for any improvements or features.