Nodes/Latent Tools/LTFloat_Step_001
ComfyUI Node

LTFloat_Step_001

A Hundredth at a Time, When You Have a Batch to Get Through

By Machines-of-Disruption·Created about a year ago·Updated 8 months ago· 27
LTFloat_Step_001
    • FLOAT
    • STRING
    value0.00

    LTFloat_Step_001 is the 0.01 member of the Latent Tools step-helper family, and it's the one you graduate to when the 0.001 sweeps start feeling like a punishment. It's a calibrated constant: you type a value, it returns the same number as a FLOAT and a STRING, with the widget stepping by 0.01.

    This family exists for sweeping a parameter across a batch of images with the seed locked - change one thing per run, compare, move on. At 0.01 per image, a 100-run batch sweeps a full unit, which is a proper range for noise parameters. The trade-off is visible: consecutive images are clearly stepped rather than smoothly interpolated, so it's better for finding the range than for building a smooth animation. Sweep coarse with this, then go fine with 0.001 once you've bracketed the interesting zone.

    How it works

    The honest version: the node does not auto-increment. It rounds the value you type to five decimals and returns it as a FLOAT and a STRING. The step is the widget's spinner increment and the label for the constant. You bump the value between runs; the node keeps it deliberate.

    The inputs and outputs

    • value - FLOAT, default 0, range ±1,000,000, step 0.01.
    • FLOAT - the value.
    • STRING - the value as text, for labeling outputs via a filename prefix.

    Wire the FLOAT into a mean/std/cfg input, the STRING into a Save Image filename, keep the seed fixed, and every image carries its own parameter in its name.

    Installing it

    Part of xl0's Latent Tools pack. ComfyUI Manager → search Latent Tools → install → restart, or:

    cd ComfyUI/custom_nodes
    git clone https://github.com/xl0/latent-tools
    

    Restart, and it's under LatentTools. The pack's only dependency is lovely-tensors (Manager installs it; manual cloners may need pip install lovely-tensors). No model downloads.

    CategoryLatentTools

    Inputs (1)

    NameTypeDefaultDescription
    valueFLOAT0.00-1000000–1000000

    Outputs (2)

    NameTypeDescription
    FLOATFLOAT
    STRINGSTRING