Painting Coder Utils
Powerful tools for image and text processing, including cross-platform workflow fixes, optimized resolution, dynamic text/image combos, and batch processing. Unlock seamless AI art creation and boost your productivity!
Nodes (20)
Collect any number of images onto one list output — ports grow as you connect
Strap as many masks as you like onto one output port
A fresh empty latent, sized the SDXL way — before you've generated anything
Empty latents in SDXL *and* Midjourney aspect ratios, batch included
The node that forces your images onto an SDXL-friendly canvas — and gives you the mask to prove it
SDXL's optimal resolutions as a dropdown, not arithmetic
SDXL and Midjourney aspect ratios, one dropdown — stop doing the math
A/B two image pipelines with one checkbox
Turn any image into a data URI — for the APIs and tools that want one
The lazy way to flip between two sampling paths without rebuilding the graph
Finally see what your mask actually looks like before it wrecks the image
Flip between two masks with a checkbox instead of a rewire
A big, comfortable text box for long prompts and captions
Dump any node's output as readable text — tensors included
Stop fighting your own pasted prompts — clean them in one node
Actually see the text flowing through your workflow
A text widget that isn't welded to the node it feeds
Glue N text inputs into one string without a pile of concatenate nodes
One checkbox to swap between two prompts
Pull an image straight off the web — URL or base64 — no browser, no save-as
ComfyUI Painting Coder Utilities

ComfyUI nodes for image sizing, text cleanup, masks and workflow repair. Choose a task below, then open the original node reference.
English | 简体中文 | 繁體中文(臺灣) | 日本語 | 한국어
Choose a task
| Task | Existing entry |
| --- | --- |
| Resize images without inventing a new workflow | ImageResolutionAdjuster |
| Create image/latent dimensions | ImageSizeCreator, ImageLatentCreator and Plus variants |
| Combine or inspect images and masks | DynamicImageCombiner, DynamicMaskCombiner, MaskPreview |
| Clean, join or display text | RemoveEmptyLinesAndLeadingSpaces, TextCombiner, ShowTextPlus |
| Route workflow values | image, text, mask and latent switch nodes |
| Load URLs or encode images | WebImageLoader, ImageToBase64 |

Existing project workflow screenshot; not an AI-generated product mockup.
Install in your ComfyUI environment
Use the Python interpreter that runs ComfyUI. Do not install a second incompatible Torch build into it. The pack declares NumPy, Pillow, Torch and Gradio dependencies in requirements.txt; review their compatibility before installing. Restart ComfyUI after cloning.
cd ComfyUI/custom_nodes
git clone https://github.com/jammyfu/ComfyUI_PaintingCoderUtils.git
# Run with the Python interpreter used by ComfyUI:
python -m pip install -r ComfyUI_PaintingCoderUtils/requirements.txt
Workflow migration and first example
The v0.3.0 namespace change can break older workflows. Back up the JSON first, use the local browser fixer to update names/path separators, then reopen it in ComfyUI. The bundled Flux workflow may need other nodes and model files; this repository does not bundle them.
Scope and permission
20 node classes are registered in init.py. These five guides do not imply five-language node UI. Full historical node settings remain in the English/Chinese references. This documentation change does not certify every ComfyUI/PyTorch version.
You may use, modify and distribute the code, including commercially, while retaining the copyright and license notice. Model, workflow-input and third-party image rights are separate. (MIT) · LICENSE