ImageBatchCopy

ImageBatchCopy takes one image from a batch and duplicates it into a new batch of quantity copies. Inputs are images, index (which frame to copy), and quantity (how many copies, minimum 2). Out comes a batch where every frame is that one image. It's the "repeat this frame" node, and it exists because batch consistency is a real requirement in ComfyUI: a lot of batch-aware operations assume every image in the tensor has the same shape and count, and if you want to inject a single frame into a batch pipeline, you need it repeated, not lonely.

The mechanics mirror ImageBatchGet: index is 1-based (1 = first image) and is clamped to the batch size, so asking for an out-of-range index silently gives you the last frame. quantity has a minimum of 2 - you literally cannot ask for a single copy; if you only need one, that's ImageBatchGet's job. The output is an IMAGE tensor of quantity identical frames, RGB or RGBA preserved as-is.

Where does this actually pay off? Real workflows:

• Seeding a batch sampler. Want to run the same reference frame through a batch of variations? Repeat it N times and wire it alongside N different prompts - every slot gets a matching image.
• Matching batch lengths. When a downstream node requires the image batch and something else to have equal counts, a quick repeat brings the image side up to match.
• Stress-testing a node. Repeat one frame into a 50-image batch and you can watch a per-frame operation handle a load without wasting generations on unique images.

Paired with its siblings - ImageBatchGet to pluck one frame, ImageBatchFork to split a batch in half, ImageBatchJoin to concatenate - it completes the small toolkit Allor gives you for reshaping batches without dropping to raw tensor hacking.

Installing it

It's part of the Allor plugin:

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

Restart ComfyUI, or install "Allor" via ComfyUI Manager. Pack-level notes: requirements.txt installs rembg and onnx for the segmentation nodes even though this node is pure tensor repetition, and Allor's repo was rebased to strip images from git history - if auto-updates start failing, the docs at nourepide.github.io/ComfyUI-Allor-Doc have troubleshooting pages for exactly that.

Real talk

ImageBatchCopy is one of those nodes you discover after manually duplicating images in an image editor to fake a batch. Once you know it exists, it's the obvious answer. The only surprise is the 2-copy minimum - if you hit it and think "I just want one," you actually want ImageBatchGet, which returns a batch of one. Otherwise: pick an index, pick a count, repeat.