ComfyUI Extension: ComfyUI_MarkdownImage
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.
This project generates an image from Markdown text using imgkit and wkhtmltoimage. It automatically scales the text to fit within the specified image dimensions.
Looking for a different extension?
Custom Nodes (4)
README
CreateMarkdownImage ComfyUI node
This project generates an image from Markdown text using imgkit and wkhtmltoimage. It automatically scales the text to fit within the specified image dimensions.
🔧 Installation
1️⃣ Install Dependencies
pip install -r requirements.txt
2️⃣ Install wkhtmltoimage
This project requires wkhtmltoimage, which is part of wkhtmltopdf. You need to install it manually.
🔗 Download wkhtmltopdf from:
👉 https://wkhtmltopdf.org/downloads.html
Make sure to add wkhtmltoimage.exe to your system PATH, or update the script with its full path.
📌 How It Works
This script:
- Converts Markdown text into HTML.
- Renders the HTML into an image using
imgkit. - Dynamically adjusts the font size to maximize readability.
- Returns the image as a PyTorch tensor.
📊 Input Fields
| Field | Type | Description |
|------------------|--------|-------------|
| md_text | String | Markdown text to be rendered into an image. Supports basic formatting (bold, italic, emojis). |
| border_color | String | HEX color or HTML named color for the image border (e.g., #FFFF00 for yellow). |
| border_size | Int | Border thickness in pixels. |
| image_width | Int | Width of the generated image in pixels. |
| image_height | Int | Height of the generated image in pixels. |
| background_color | String | HEX color or html color name (e.g., "lightblue"). |
🚀 Usage
from create_markdown_image import CreateMarkdownImage
generator = CreateMarkdownImage()
image_tensor = generator.createmarkdownimage(
md_text="**Hello World!** 🚀",
border_color="#000000",
border_size=2,
image_width=800,
image_height=200,
background_color="white"
)
This will generate an image containing "Hello World! 🚀" with a black border and a white background.
🛠️ Troubleshooting
- If
imgkitfails, ensurewkhtmltoimage.exeis installed and accessible. - If text doesn't scale correctly, adjust the
image_widthandimage_heightfields.
📜 License
This project is licensed under the MIT License.
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.