Make Image List

You've got a handful of separate images sitting on the canvas - from different generations, different branches, wherever - and you want a downstream node to process each one individually rather than as a stacked block. Make Image List takes loose image connections and packs them into a genuine ComfyUI list, not a batch. That distinction is the whole reason this node (and its sibling, Make Image Batch) exists as two separate tools instead of one.

If the batch-vs-list difference is new to you: a batch is many images bundled into a single tensor and processed together in one shot; a list is separate items that ComfyUI re-runs the receiving node on, once per item. A KSampler with batch size 4 hands you a batch. A detailer you want run per-image, a save node you want firing per-file, anything you want looped over one item at a time - those want a list. Make Image List is how you build that list from individual image wires you already have, rather than from an existing batch (that's Image Batch to Image List's job - splitting, not assembling loose connections).

How it works

The schema shows a single optional image1 slot - the growing-slot pattern Impact Pack uses across its "Make ... List/Batch" family. Connect an image and a second slot appears; connect that and a third shows up, and so on. There's no fixed cap; the node grows to however many sources you feed it. On execution, everything you connected comes out as one ComfyUI list, in the order you wired them.

The inputs and output

• image1 - the first image slot (optional in the schema, but connecting something here is what starts the chain and reveals the next slot). Keep connecting until you've added every image you want in the list.

Output is a single IMAGE, explicitly flagged as a list. Anything downstream wired to it will run once per item.

How to install it

Via ComfyUI Manager: search ComfyUI Impact Pack, install, restart. Manually:

cd ComfyUI/custom_nodes
git clone https://github.com/ltdrdata/ComfyUI-Impact-Pack comfyui-impact-pack
cd comfyui-impact-pack
python -m pip install -r requirements.txt   # ComfyUI's python; python_embeded on portable

then restart. No models - a lightweight structural node, zero GPU cost of its own.

Common issues & troubleshooting

Make Image List vs Make Image Batch - which do I want? Use Make Image List when you want each image processed separately downstream (a detailer run per image, individual saves). Use Make Image Batch when you want them treated as one combined unit (a single batched sampler pass) - and note that Batch also reconciles differing image sizes for you, which List does not need to, since list items stay independent.

Only one image came through. You likely only connected the first slot. New slots only appear after the previous one is filled - keep wiring until every source image is connected.

Downstream node ran once on everything instead of once per image. That means it received a batch somewhere along the way, not the list this node produced - check that nothing between this node and the consumer accidentally converted it back (an Image List to Image Batch inserted by mistake, for instance).

Slower than expected. List execution genuinely re-runs the downstream chain once per item, which is correct for per-image work but costs more total time than a single batched pass. If the downstream node can handle a batch natively and you don't need per-item behavior, Make Image Batch is the faster choice.