Just a pack of different LoRA's that will break your works and add glitch art distortions. I wanted to make a LoRA, that would work even without activation words, and so did I. BUT! If you add some actiovation words like 'glitch', 'distortion', 'glitch art style' the style appears to become more intense

The LoRA works with any weight parameters, but i prefer 1.25-1.5, if you go higher, image is still cool sometimes, but more like some strange abstract thing :D

Description of models (as i see em):

Diferent models are not just iterative improvment, they are more like variations. Check comparison table above and get the one that you like the most

v1 - generic digital glitch

v1.0.1 - digital glitches, but "softer" effect

v1.0.2 - more agressive style, was trained on more diverse dataset, tend to make slitscan stuff

---

You can support me and request new lora's/checkpoints at: https://boosty.to/smgbuddies/donate

描述:

训练词语: