A Random Wrong Strange Funny LoRa
This is the typical LoRa that you train and it goes wrong. The time comes to test it, and it is not at all what you were looking for, and not only that, it ruins all the images. You save it in a dark corner of your hard drive and promise to test it again another day more seriously. That never happens.
Until one day, someone sees something in a failed image generated with that LoRa and asks you about it. Then you look for it and test it again. It's still bad, but it does things you didn't expect and you spend an entire afternoon laughing at it. What are you doing with that?
You publish it.
I want to see all your failed images please!!! Fill this wall please xD.
If you like my work and want to support me, or are interested in information about my upcoming projects or updates to my models, visit: https://ko-fi.com/devildonia.
Description
v3.0
FAQ
Comments (3)
I created a review for this amazing lora, but I don't know how to make an image as the cover image, the arrange functionality seems not working, can anyone help me?
It appears to be a LORA that "corrects or further displaces a specified prompt" under the guise of a LORA that "corrects or further displaces a specified prompt".
I've been testing it thoroughly for a while, and I still don't understand what exactly it does. When you seem to have understood how it works, changing a parameter and it does completely different things. you go up steps: it changes. You switch sampler: it does different things. What about hires. Fix: It does different things again. It alters the checkpoint in different ways and very subtly or very exaggeratedly.