🏆 EveryoneNobel

| Overview | Requirements | Quick Start | Contributors |

Overview

EveryoneNobel aims to generate Nobel Prize images for everyone. We utilizes ComfyUI for image generation and HTML templates to display text on the images. This project serves not only as a process for generating nobel images but also as a potential universal framework. This framework transforms the ComfyUI-generated visuals into final products, offering a structured approach for further applications and customization.

We share how we build the entire app and sell the product in 30 hours in this blog here.

You could generate the picture without text here.

Requirements

1. Install ComfyUI

Follow the instructions in ComfyUI repo to install ComfyUI. Open ComfyUI and install the missing custom nodes and models for workflow in resources/workflow/nobel_workflow_for_install.json. Our lora model is here.

2. Install requirements

# cd to EveryoneNobel main folder
npm install
pip install -r requirements.txt

3. write .env (optional when using main_without_openai.py)

Create a .env file in the main folder with the following content:

API_KEY=YOUR_OPENAI_API_KEY

Quick Start

1. Start ComfyUI server

An example for starting the server.

# cd to ComfyUI main folder
{ComfyUI_python} main.py --port 6006 --listen 0.0.0.0

(NEW) The {ComfyUI_python} is the path to the python inside your ComfyUI. For me, it is the python.exe under python_embeded folder in ComfyUI.

2. Run main.py

An example

python main.py \
  --name "somebody" \
  --subject "2024 nobel prize" \
  --content "Do nothing" \
  --image_path "resources/test/test.jpg" \
  --comfy_server_address "127.0.0.1:6006"

Parameter Explanations:

• --name: The name of the individual.
• --subject: The subject of the prize.
• --content: Description contribution of the individual. (AI will use this to generate the text at the bottom of the image)
• --image_path "resources/test/test.jpg": The file path of the input image
• --comfy_server_address "127.0.0.1:6006": Sets the address of the ComfyUI server that will handle the image generation.

(NEW) Example for not using openai api to generate text:

python main_without_openai.py \
  --name "somebody" \
  --subject "2024 nobel prize" \
  --content "Do nothing" \
  --image_path "resources/test/test.jpg" \
  --comfy_server_address "127.0.0.1:6006"

Contributors

Star History