Integer List Generator | akatz-loops

If your workflow needs a run of integers - a list of seeds, a sweep of values, a batch of step counts - IntegerListGeneratorNode produces the whole sequence in one go. You give it a starting integer, a quantity, and a mode, and it emits a list of INTs you can feed into index pickers, samplers, or a loop. It's the pack's way of saying "don't type ten numbers by hand."

Three inputs, and the first two are self-explanatory:

• start_int (default 0, up to 4294967295) - where the sequence begins.
• quantity (default 5, up to 10000) - how many integers to generate.
• control_mode (increment or random) - the fun one.

In increment mode it's a straight arithmetic run: start_int, start_int+1, start_int+2, … for quantity entries. In random mode it's a batch of random integers in the full 32-bit range - and here's the genuinely useful detail: it uses start_int as the master seed. The sequence is deterministic. Same start_int, same list every run; change start_int, get a different reproducible set. That's exactly what you want for a seed sweep: a fixed "seed A set" and a "seed B set" you can hand to people and have them reproduce, all without a separate randomizer node.

What you'd actually do with it

The workhorse pattern is parameter sweeps. Generate a list of seeds, pull one per run with GetIntFromList, and run the workflow once per seed - you get variation without touching the graph. Pair it with this pack's loops and you can iterate over the list inside one run: generate the values, pick by the loop's current iteration index, and each pass uses the next integer. That combo - generator → index picker → loop - is the pack's version of a for-each over a numeric list, and it's why this node sits in the Lists category rather than being a pure utility.

It's also worth flagging what it is not: it's not a noise/seed node in the sampling sense. The output is a list of integers, and those integers are whatever you say they are. If you want them to behave as seeds, you feed them into a node that consumes an INT as a seed. The node doesn't know or care.

Install

Part of Akatz-Loop-Nodes (akatz-ai). Via ComfyUI Manager, search "Akatz-Loop-Nodes"; or:

cd ComfyUI/custom_nodes
git clone https://github.com/akatz-ai/Akatz-Loop-Nodes

Restart ComfyUI. Pack dependency is opencv-python (image nodes), no models. The custom_nodes.ComfyUI-Execution-Inversion module path on listings is lineage from BadCafeCode's execution-inversion demo - not a separate install.

The one gotcha: in random mode the start_int range also limits your seed range only in the sense that a 32-bit value fits the standard seed space - if you pick a start_int near the max, the generated randoms are still full-range, so you're fine. And remember quantity is capped at 10000; a "list" of 100k ints isn't happening here, which is honestly a mercy for your RAM.