Convert Greyscale

Sometimes you want a black-and-white version of an image mid-graph - feeding a preprocessor that expects greyscale, doing a quick contrast/luminance check, or just producing a stylized output - and don't want to break out an external editor to do it. ConvertGreyscaleNode is exactly one thing: image in, greyscale image out.

It's part of ComfyUI-LogicUtils, a small utility pack by GitHub user aria1th - publicly AngelBottomless, the trainer behind Illustrious XL, the Danbooru-trained SDXL fine-tune that displaced Pony Diffusion as the open-source anime default. LogicUtils itself has nothing to do with that model training directly; it's a general-purpose bag of glue nodes - image ops, dict manipulation, type conversion - the kind of thing that accumulates around someone who does a lot of automated batch image processing. The pack is honest about being under-documented: the README notes "proper documentation is being prepared, however there are too many nodes," which is a fair description of a lot of the smaller, self-explanatory nodes in it, this one included.

How it works

There's genuinely nothing to configure here - the node takes the incoming image and converts it to greyscale, returning it as a standard IMAGE output. No parameters to tune, no direction or method choice like its Concat* siblings in the same pack. If you need finer control over how the greyscale conversion is weighted (luminance-based versus a flat average, for instance), this node doesn't expose that - it's a fixed, single-purpose operation, not a tunable filter.

The inputs and outputs that matter

• image (IMAGE, required) - the image to convert. That's the entire input surface.
• Output: a single IMAGE - the greyscale result, ready for a Preview Image, Save Image, or any downstream node.

How to install it

Via ComfyUI Manager: search ComfyUI-LogicUtils, install, restart. Or by hand:

cd ComfyUI/custom_nodes
git clone https://github.com/aria1th/ComfyUI-LogicUtils

Restart ComfyUI once it's in place. No model files needed - this is a plain image-processing operation with no checkpoint or download involved. The README mentions an opt-in auto-install hook (COMFYUI_LOGICUTILS_AUTO_INSTALL=1, with COMFYUI_LOGICUTILS_SKIP_INSTALL=1 to disable it) for other parts of the pack; a single-purpose node like this one doesn't need it.

Common issues & troubleshooting

Output still looks like it has some color. Depending on your viewer, a greyscale image saved in an RGB format can look "off" rather than obviously black-and-white at a glance - check the actual pixel values or view it in an image inspector rather than trusting a thumbnail preview at a glance.

You wanted control over the conversion method and don't see an option. There isn't one on this node - it's fixed-behavior by design. If you need a specific luminance formula or channel-weighting, you'll need a different node (or a Color Correct-style node from another pack) rather than expecting a hidden parameter here.

Can't find the pack in Manager's search. It's a low-visibility repository - if the search doesn't surface it, use Manager's "Install via Git URL" option and paste https://github.com/aria1th/ComfyUI-LogicUtils directly.

No response to a bug report. This is a personal side-project pack, not something with an active support cycle - its author's public repository activity moved away from image-generation tooling entirely in early 2026. For a node this simple, though, that's unlikely to matter much; there's very little surface area for something to break.