🖼🖼 Merge Images/Videos 📹📹 (Vertically)

The vertical twin of the pack's horizontal merge node: stack two to four images on top of each other into one tall image, instead of side by side. Same job, different axis - the kind of node you don't think about until you need it, and then you're glad it exists instead of hand-rolling an image-concat step.

How it works

Wire in two to four images and they get stacked top to bottom into a single output. There's no built-in resizing - it's a plain vertical concatenation, so images of mismatched width will look wrong (or error) unless you resize them to a common width upstream first.

Like its horizontal sibling, this works on IMAGE-typed batches generically, which means it doubles as a video-stitching node, not just a stills tool - the author built both of these with video comparisons in mind. The two are meant to be used together: horizontal-merge two pairs of images, then vertical-merge those two combined rows, and you've built a 2x2 grid without touching a compositing tool outside ComfyUI. That's genuinely handy for eyeballing four sampler/scheduler combos, or four denoise strengths, at a glance.

The inputs and outputs that matter

• image1, image2 - required, the two images stacked at minimum.
• image3, image4 - optional, extend the stack to three or four.
• Output: one IMAGE, everything stacked top to bottom in the order you wired them in.

How to install it

Through ComfyUI Manager: search Bjornulf_custom_nodes, install, restart ComfyUI. Manually:

cd ComfyUI/custom_nodes
git clone https://github.com/justUmen/Bjornulf_custom_nodes

cd into the folder, run pip install -r requirements.txt (a venv keeps this tidy relative to your other custom node packs), and restart. Windows portable users should run the equivalent pip install command through the python.exe inside python_embeded, not a system Python install.

Common issues

Width mismatches are the practical gotcha - if your images came from different aspect ratios or different upscalers, resize them to a common width before stacking, since this node isn't doing any scaling for you.

And the same caveat applies here as everywhere else in this pack: it lives inside one Python package alongside much heavier nodes (Ollama, TTS, the CivitAI and FalAI API generators), all sharing a single requirements.txt. If this node is missing from your search entirely rather than just erroring, that usually means something else in the pack failed to import - check the ComfyUI startup console for a traceback before assuming this specific node is the problem.