FL Image List To Image Batch

This one fixes a specific, maddening ComfyUI plumbing problem. ComfyUI has two different ways of carrying multiple images: a list (where the graph runs once per item) and a batch (a single tensor with N images stacked in it). They look similar, they are not interchangeable, and sooner or later a node hands you a list when the next node wants a batch - and everything breaks with a confusing error. FL_ImageListToImageBatch is the adapter: list in, batch out.

If you've never hit this, count yourself lucky and keep this node in your back pocket. If you have hit it - that moment where a save node or a video-combine node chokes on what looks like perfectly good images - this is the fix.

Why lists and batches differ

A batch is one tensor of shape [N, height, width, channels]: N images that all share dimensions, processed together. A list is N separate items that ComfyUI feeds through the graph one at a time. Samplers, video nodes, and grid nodes generally want a batch (one tensor to operate on). Some nodes - directory loaders, splitters, per-item processors - emit a list. When the two meet, you need a bridge, and this is it.

How it works

The node collects the images from the incoming list and stacks them into a single batched tensor. If the images aren't all the same size (a list can hold mismatched dimensions where a batch cannot), it resizes them to match the first image using Lanczos interpolation, so the stack is valid. Out comes one clean batch.

The inputs and outputs

• images - the list of images to combine. That's it.

Output is a single IMAGE - a batched tensor you can feed to anything that expects a batch: a sampler, a save node, FL_ImageBatchToGrid, a video combiner. It's the exact inverse of the pack's FL_ImageBatchToImageList, which goes the other way.

How to install it

Part of filliptm's Fill-Nodes pack. ComfyUI Manager → search ComfyUI_Fill-Nodes → Install → restart, or:

cd ComfyUI/custom_nodes
git clone https://github.com/filliptm/ComfyUI_Fill-Nodes
pip install -r ComfyUI_Fill-Nodes/requirements.txt

No model, no dependencies of note.

Common issues & troubleshooting

It seems to do nothing / errors on a batch input. This node is for converting a list. If what you're feeding it is already a batch, it's the wrong tool - you don't need it, and depending on how the upstream is typed it may not behave. The tell is upstream: was the previous node's output a list (the socket often behaves per-item) or a batch? Only reach for this when it's genuinely a list.

Images came out resized / distorted. Because a batch requires uniform dimensions, mismatched images in the list get resized to match the first one. If the first image is a different aspect ratio than the rest, everything conforms to it and can stretch. Put your reference-size image first, or normalize sizes before this node so nothing gets squished.

Which direction do I need? If a node wants "a batch" and you have separate/per-item images, use this (list → batch). If a node needs to process images one at a time and you have a batch, use the inverse, FL_ImageBatchToImageList (batch → list). Getting the direction right is 90% of solving list/batch errors.