Ard Basic Load Image

ComfyUI's built-in Load Image hands you two things: the image tensor and an alpha mask. That's it. If you want to know the image's dimensions - to set an empty latent, to check aspect ratio, to feed into img2img - you'd normally add a whole separate "Image Size" node. Ard Basic Load Image is the built-in node plus those dimensions, in one place. It returns image, mask, width, height, and image_path.

Everything you already know about Load Image applies, because this is basically a copy of it: the image dropdown lists files in your ComfyUI/input folder (with the upload button), it handles EXIF orientation, it can load multi-frame files (each frame becomes a batch item), and it converts to a 0–1 float tensor plus a mask where an alpha channel exists. The additions are the two INT outputs (width, height in pixels) and the image_path STRING output (the full annotated file path, which is handy if another node wants to reference the file directly).

Inputs: just image. Outputs: image, mask, width, height, image_path. That's the whole interface, and the honest pitch is exactly the width/height pair - in img2img or the pack's Control Box world, you almost always want the loaded image's dimensions as numbers, and this saves you a node and a wire.

When would you reach for it? Two classic spots:

• img2img: load a reference, use its width/height to set the empty latent you're about to denoise, so you don't accidentally resample to a different aspect ratio.
• Compositing: you want the size to drive a canvas or positioning math downstream.

Gotchas are inherited from Load Image more than invented: the image has to live in the input folder (or you upload it); a 64×64 zero mask is returned when the image has no alpha channel, same as the core node - so don't assume a "real" mask. And note this node reads the file at execution time, so replacing the file and re-running picks up the new one (that's true of any Load Image).

Honest verdict: it's a genuinely reasonable small-quality-of-life node, and one of the more defensible ones in this pack. It doesn't add a feature - it fuses two nodes you'd use anyway. If you load images often and always chase their dimensions, this saves you a node every time. If you never need the size, there's zero reason to prefer it over the core node.

Install is the standard pack route - ComfyUI Manager → search ComfyUI-Ardenius, or:

cd ComfyUI/custom_nodes
git clone https://github.com/ArdeniusAI/ComfyUI-Ardenius

then restart ComfyUI. Pack-wide notes: civitai and moviepy install with the pack (irrelevant to this node), and a single import error in the console at startup is the Save Image node wanting comfyui_controlnet_aux - the rest of the pack, including this node, loads fine.