Input: Integer (lab)

Input: Integer (lab) holds a whole number (default 0) and outputs it as an INT. It's the integer member of ComfyLab's trivial-but-useful input trio, and its reason to exist is the same as the float and boolean versions: ComfyUI has plenty of integer widgets buried inside nodes, but sometimes you need an integer value on a wire.

The most concrete use is batch sizing. Seeds, step counts, and batch sizes are integers, and when a queue is feeding them in one at a time, the value has to come from somewhere typed. List: Random Seeds (lab) produces integers for seeds; an InputInteger produces the single one you want to reuse across many places.

What it is

One widget, value (an integer, default 0), one output, integer. No min/max, no step concerns - just a typed whole number.

Where it earns its keep:

• Step count or batch size as a shared constant - set it once, feed a KSampler's steps and a format string's {steps} placeholder from the same source.
• Index math - offsets, counts, loop limits feeding the pack's list nodes (List: Limit's nb_elements, for instance) or pagination values into the XY plot queue.
• Seed reuse - a fixed seed you want visible at the top of the workflow instead of tucked inside a sampler.

Install

ComfyLab Pack installs as one package (nodes display with "(lab)" suffixes). ComfyUI Manager: search ComfyLab Pack, install, restart. Manual:

cd ComfyUI/custom_nodes
git clone https://github.com/bugltd/ComfyLab-Pack.git
cd ComfyLab-Pack
pip install -r requirements.txt

Restart after. No model files, light requirements.

Gotchas

• Integers only. A 2.5 belongs in Input: Float; forcing it through here will error or silently truncate downstream.
• It's a constant, not a spinner with semantics. No min/max guards, no "must be a multiple of 8" validation - that's your job (or a multiple_of knob elsewhere in the pack).

If you squint, these input nodes are just "widgets you can wire." That's the entire trick, and it's exactly what makes a graph where the numbers you care about are visible and shared instead of scattered across fifteen node internals.