Image List To Batch ♾️Mixlab

The mirror image of ImageBatchToList_, and just as simple: once you've processed a list of images one at a time - each item re-running the downstream chain individually - this node collapses that list back into a single batched IMAGE tensor, for whatever comes next that expects a batch rather than a per-item list.

How it works

Feed it a list of images and it stacks them into one batch. That's the whole node. It's the step you insert right after a per-item processing loop and right before something that wants one tensor - a grid/contact-sheet node, a batch-aware upscaler, a video-combine node that expects all frames as a single batched input rather than N separate executions.

The inputs and outputs that matter

• images (IMAGE, required, taken as a list) - the images to recombine.

The single output is IMAGE, explicitly not a list this time - a single batched tensor.

Its companion, ImageBatchToList_ (same pack), does the opposite conversion - splitting a batch into a list for per-item execution.

How to install it

Ships with the Mixlab pack.

• ComfyUI Manager: search comfyui-mixlab-nodes, install, restart.
• Manual: cd ComfyUI/custom_nodes && git clone https://github.com/shadowcz007/comfyui-mixlab-nodes.git, install requirements (install.bat on Windows, or ../../../python_embeded/python.exe -s -m pip install -r requirements.txt), restart.

No model downloads - a pure list/batch conversion utility.

Common issues & troubleshooting

Shape mismatch error when stacking. All the images in your list need matching dimensions to stack cleanly into a single tensor. If your list came out of a pipeline where individual items got resized inconsistently (a per-item crop with different results, say), fix the sizing before this node - resize or pad everything to a common size upstream, not after you've already tried to batch it.

Not sure whether I need this node at all. If you're not doing per-item processing that required splitting into a list in the first place, you may never need ImageBatchToList_/ImageListToBatch_ at all - a lot of workflows are fine staying in batch form the whole way through.

Order of images in the batch looks wrong. The batch should preserve the list's order, so if the sequence is scrambled, check whatever produced the list upstream rather than assuming this node reordered anything.

Worth remembering: this pair of nodes (with ImageBatchToList_) is glue, not a feature - it exists purely to bridge ComfyUI's two execution modes for images. If you never split into a list in the first place, you'll never need this half of the pair either. The only reason to reach for it is that something upstream in your graph forced per-item execution and you now need a single tensor again for whatever comes next.