Image Batch To Image List

This node exists because ComfyUI has two ways of holding multiple images and they are not the same thing, and mixing them up is one of the most common quiet mistakes in the whole ecosystem. easy imageBatchToImageList converts a batch into a list. That one conversion unblocks a surprising number of "why is this node only processing one image?" problems.

Batch vs. list, quickly

A batch is a single tensor with a batch dimension - all the images stacked together, all the same size, processed in one pass. A list is separate items that ComfyUI feeds through the graph one at a time, re-running each downstream node per item.

Why does it matter? Some nodes only act on the first image of a batch, or produce one output for the whole batch. If you want a node to genuinely run on each image - caption every frame, upscale each one, save them individually with distinct filenames - you often need a list, not a batch. That's when you drop this node in.

How it works

It takes the batch tensor and unstacks it along the batch dimension into individual list items. Downstream nodes then execute once per item automatically.

The inputs and outputs that matter

• image - the batched IMAGE to split.
• IMAGE (output, marked as a list) - the same pictures, now as separate list items. Wire this into the node you want to run per-image.

No settings. It does one thing.

How to install it

ComfyUI Manager: search ComfyUI-Easy-Use, install, restart. Manually:

cd ComfyUI/custom_nodes
git clone https://github.com/yolain/ComfyUI-Easy-Use

then install.bat / pip install -r requirements.txt, restart. No models.

Common issues & troubleshooting

Everything downstream now runs N times. That's the feature, not a bug - a list re-executes the graph per item. If that's slow or not what you wanted, you probably wanted the batch after all. Convert back with easy imageListToImageBatch.

"Which one do I need?" heuristic. If a node is only touching the first of several images, or you need per-image filenames, you want a list (this node). If a sampler or model wants everything at once for throughput, you want a batch. When in doubt, this pair - imageBatchToImageList and its inverse imageListToImageBatch - lets you flip between them wherever a node is being picky.

Preview/save shows only one. Some save/preview nodes collapse a batch to a single displayed image; converting to a list makes them emit each one. This is the usual reason people go looking for this node.