Simple Expression Integers

ComfyUI's built-in math primitives cover addition, subtraction, and the occasional clamp - and that's about it. The moment you want (a + b) * 2 or a step count derived from two other values, you're stacking nodes or reaching for a custom script node. Simple Eval Integers is the honest fix: a text box where you type the expression, three integer inputs (a, b, c) to plug values into it, and a result that's exact.

It's one of three siblings in the pack - the float and string variants do the same thing with different input types - and it's the one you reach for when your numbers are whole. Think: computing a batch size from a count, deriving a latent dimension, or turning a slider into a different slider.

How it works

Under the hood it's simpleeval, a deliberately restricted Python-expression evaluator. That's the whole trick and it's a good one: you get real expressions - (a + b) * 2, a // c, max(a, b) - but there's no import os or arbitrary code execution hiding in there. It's safe to run inside a workflow that executes untrusted-ish values.

Type your expression in python_expression, wire a/b/c (or leave them at 0), and the node evaluates. print_to_console optionally echoes the result to the terminal - handy when you're debugging why your math is producing nonsense.

The outputs are the interesting part, because they're cast three ways:

• INT - integer result
• FLOAT - float result
• STRING - string representation

That last one is the sleeper feature. Wiring the INT into a numeric input is obvious, but the STRING output lets you build dynamic text - a "2 of 10" label or a filename fragment - from the same computation. One node, three types, no conversion nodes.

Install

Part of ComfyUI Text Processor:

cd ComfyUI/custom_nodes
git clone https://github.com/rookiestar28/ComfyUI_Text_Processor.git
pip install -r requirements.txt

Restart after. Manager: search "ComfyUI Text Processor". simpleeval is in the pack's requirements, so it comes along automatically.

Gotchas

Division on integers behaves like Python: 5 / 2 gives 2.5 in the FLOAT output but 2 in the INT output (truncation, not rounding) - if you expect rounding, write round(a / b) yourself. And simpleeval is a restricted subset of Python, so fancy syntax like list comprehensions or lambdas will just error. Keep expressions to arithmetic, comparisons, and the built-in functions it knows, and it'll behave. For decimal math where truncation matters, use the float sibling instead - this one is for when the result should be whole.