Load Image (Yogurt Nodes)

The honest review of YogurtLoadImage is: it does what the core Load Image node does, with the same dropdown, the same IMAGE output, and the same alpha-to-MASK behavior. If you loaded it expecting something novel, you'll be disappointed. If you loaded it because you're building a workflow inside the ComfyUI-YogurtNodes ecosystem and want everything to behave uniformly, you'll be quietly relieved.

That context matters, because this pack - yogurt7771's 150+ node all-in-one, MIT-licensed, auto-generated README, zero community buzz - is a grab-bag. Some nodes are genuinely novel (the LoRA surgery suite, the mask region planner). Some are re-exports of things core ComfyUI already has. Load Image is the latter, and that's a feature: when a pack ships its own loader, files load the same way across every graph you build with it, and you don't need to remember which node in the pack expects images from which source.

How it works

One required input, image, which is an enum populated from your ComfyUI input directory - drop a file into ComfyUI/input and it appears in the dropdown. Two outputs:

• IMAGE - the image tensor, standard format, wired into samplers, VAEs, and everything downstream.
• MASK - the alpha channel as a mask tensor. Transparent pixels become the mask; a fully opaque image produces an empty mask. This is the same behavior as core ComfyUI's LoadImage, useful for working with PNGs that carry alpha.

When to reach for it

• Any workflow that already uses other Yogurt Nodes, for consistency.
• When you specifically want the alpha channel as a mask alongside the image.
• When you're following a workflow someone shared that was built around this pack.

When not to bother: if you're just loading a normal image into a vanilla workflow, core LoadImage does the identical job and you already have it. Don't install a whole pack for a re-export. Install it because you want the rest of the pack's list, logic, and LoRA tools, and this node conveniently rides along.

Install

ComfyUI Manager, search ComfyUI-YogurtNodes, install, restart. Or:

cd ComfyUI/custom_nodes
git clone https://github.com/yogurt7771/ComfyUI-YogurtNodes
cd ComfyUI-YogurtNodes
pip install -r requirements.txt

Restart, look under "Yogurt Nodes". Light dependencies (numpy, pillow, requests, opencv-python), no model downloads. The one habit to keep: files must live in ComfyUI/input to show up in the dropdown - if your image is elsewhere on disk, the pack's YogurtLoadImagePath takes an absolute path instead.