power

Raising a number to a power is the operation behind every "doubling" calculation - 2^n, growth curves, resolution doubling chains - and it's the kind of thing a generic math node makes feel heavier than it is. power from the Basic data handling pack is a clean integer exponent node: base in, exponent in, integer out. If your workflow ever computes 2^x or similar, this is the node that makes it a one-liner instead of a formula-string puzzle.

What it does

base ** exponent, plain Python exponentiation. 2 ** 10 is 1024. 3 ** 3 is 27. Two details worth internalizing:

• An exponent of 0 gives 1 - the default, and correct, since anything to the power of zero is one. That default means double-clicking the node gives you 1, not an error.
• Keep exponents non-negative. Python's int ** negative-int returns a float - 2 ** -1 is 0.5 - and a float doesn't fit this node's INT output cleanly. For fractional powers, use the pack's float power node instead.

Inputs and outputs

• base - INT, default 1. The number being raised.
• exponent - INT, default 0. How many times to multiply it.
• Output INT - the result of base ** exponent.

Install

Part of the Basic data handling pack. ComfyUI Manager → search "Basic data handling" → install → restart. Or:

cd ComfyUI/custom_nodes
git clone https://github.com/StableLlama/ComfyUI-basic_data_handling

then restart ComfyUI. Pure Python, no dependencies, no model downloads.

When it's the right tool

The classic use is resolution math: a "tile until it's big enough" loop that squares dimensions, or an upscale factor computed as 2^n for N refinement passes. Pair it with IntMultiply for anything more complex than a plain power. And the growth warning: exponentiation gets big fast, and unlike fixed-width languages Python won't stop you - 10 ** 100 is a 101-digit integer that will happily flow into the next node. Great for exploration, but put a cap on it when it's feeding real parameters.