Sign In

Subatomic

0

Verified:

SafeTensor

Type

LoRA

Stats

70

0

Reviews

Published

Feb 16, 2025

Base Model

SDXL 1.0

Training

Steps: 2,425
Epochs: 10

Usage Tips

Clip Skip: 1
Strength: 1

Trigger Words

subatomic

Hash

AutoV2
B616191485

The description I made of this model in the proceding paragraph is accurate, but it's a bit of an undersell. I didn't fully understand how impactful my strategy for creating this model was when I created it. This model was created to make blacklight poster styled images. 😄

You'll notice the sample images don't reflect that. That is because I discovered this model hits well above its intended design due to the way it was created. I've since learned exactly why it does what it does. It will help you draw just about anything non-photorealistic.

This model can add texture and movement to your images. This effect can be subtle or in-your-face, depending on your prompt and the image subject matter. You might start your prompting with dark_background or black_background but that is not a necessity. All sample images I have included were made with this model at 1.0 strength.


I've included below some of the more common tags I used in the dataset as "secondary triggers".

Trigger: subatomic

*Clip Skip (important note) use clip 1 or 2. Which you might choose will be a matter of taste, but occasionally the checkpoint you are working with will determine which to choose. Experiment with both.



Positive Prompts (secondary triggers)

black_background - dark_background - no_humans - glowing - orb - magic_circle - light_particles - scenery - fantasy - star_(sky) - magic - water - abstract - space - planet - crystal - ripples - energy_ball - fire - still_life - sun - colorful - reflection

Positive Prompt Ideas

swirl - diffraction_spikes - ribbon - double_helix

Clip Skip

1 or 2

Hide