Image Chunk ⚡🅡🅞🅣🅘

Worth being upfront about this one: it's filed in the pack's Utility/Batches category - the same family as ImageIndexSelect, which pulls a single frame out of a batch by position - and "chunk" strongly implies splitting a larger image batch into smaller groups. But the schema exposed for this node is unusually thin: a single required IMAGE input and a single IMAGE output, with no chunk-size, chunk-count, or overlap parameter documented anywhere in it. Rather than guess at widget names and ranges that aren't backed by anything, here's what's actually confirmed, plus what to check once it's installed.

What we know for sure

The node takes an IMAGE batch and returns an IMAGE. That's the entirety of the documented interface. Given the category it sits in and the "chunk" name, the most plausible behavior is that it divides an input batch into smaller sub-batches - the kind of operation you'd reach for to break a long video sequence into manageable pieces for chunked processing, or to keep memory usage down by not running every downstream node against a huge batch at once. But any actual chunk-size control, if one exists, isn't captured in the info we have to work from here.

What to do about it

If you install this node and don't see an obvious "chunk size" or "count" widget on it, it may genuinely only expose the one visible in ComfyUI's own node inspector once installed - check the node directly in your own ComfyUI instance rather than assuming it matches a typical chunking node's interface. If it does the batch-splitting job the name suggests, it likely needs a second output (or a LIST/looped consumer downstream) to actually receive the separated chunks - worth checking what outputs the node actually shows once it's on your canvas, since a "chunk" operation returning a single IMAGE type is at minimum unusual for the concept.

Installing it

Via ComfyUI Manager: search "RyanOnTheInside," install, restart. Manually:

cd ComfyUI/custom_nodes
git clone https://github.com/ryanontheinside/ComfyUI_RyanOnTheInside.git
cd ComfyUI_RyanOnTheInside
pip install -r requirements.txt

then restart. Given the category (a plain utility node, not one of the Flex/audio/particle families), it's unlikely to need anything beyond the pack's base dependencies - no model downloads expected.

Common issues & troubleshooting

Can't find a chunk-size setting. That matches what's documented here - this node's confirmed interface doesn't show one. Look at what widgets ComfyUI itself renders once the node is on your canvas; that's more reliable right now than anything written about it secondhand.

Unsure this is the right node for splitting a batch. If you specifically need to slice a batch into indexed pieces rather than a full chunking operation, ImageIndexSelect - this pack's sibling node in the same Utility/Batches category - is a confirmed, well-documented way to pull individual frames by position, which may cover your actual need more predictably while this node's exact behavior gets clarified.