Multiple Image Blend 2

This is the pack's original Multiple Image Blend with one addition: an align control. That single extra field exists to solve the most obvious limitation of the original - plug in images of different resolutions and this version has an actual answer for what to do with them, instead of erroring or producing a broken composite.

What changed, and why it matters

The base Multiple Image Blend node has no concept of size mismatch at all - it assumes every connected image is already the same resolution. The moment your sources aren't (a base render at 1024×1024 and an overlay texture at 512×512, say), you're stuck pre-resizing everything by hand before the blend will even work. This node's align field exists specifically to remove that step: it gives you three ways to reconcile differently-sized inputs - resize them to match, or anchor them without resizing, at top-left or centered.

Beyond that one field, the node behaves exactly like its predecessor: one required base image, up to three optional additional layers, and one shared opacity value governing how strongly they blend together.

Inputs and outputs that matter

• image_a (required) - your base image, the only image input the node insists on.
• opacity (required, 0–1, default 1, step 0.01) - one shared strength value across whatever additional images you've connected.
• align (required, dropdown: size, top left, center) - how mismatched image sizes get reconciled. size most likely resizes layers to match the base; top left and center anchor smaller or larger layers at that position instead of resizing. If your source images are already the same resolution, this setting won't visibly change anything.
• image_b, image_c, image_d (all optional) - extra layers, same as in the original node.

The output is a single image.

Installing it

ComfyUI Manager: search ComfyUI-ReplenishNodes, install, restart. Or manually:

cd ComfyUI/custom_nodes
git clone https://github.com/hben35096/ComfyUI-ReplenishNodes

No model downloads or extra dependencies documented for this node.

Where people get tripped up

Picking the wrong align mode for the job. If your layers genuinely need to fill the same frame (an overlay meant to cover the whole base image), size is what you want. If you're compositing a smaller element onto a larger canvas and want it placed rather than stretched, top left or center will behave very differently from each other depending on your image's actual content - there's no live preview of the anchor point, so it's worth testing on a throwaway run before committing to a long batch.

Reaching for this when the original would do. If all your source images are already the same size, the extra align field does nothing useful - the plain Multiple Image Blend node is functionally identical and one field simpler to configure. Use this version specifically when size mismatch is the actual problem you're solving.

Same shared-opacity limitation as the original. One opacity value still governs every connected layer together - this node adds size handling, not per-layer strength control.