Hello welcome!
Here I leave a small introduction of the LoRA:
I couldn't really decide whether to delete the epoch, the remaining nine are very good, but I just wanted to create a concept out of interest, you can use it however you want, whether on t-shirts, hoodies, stickers or whatever you like just show your designs here.
Well, leaving that aside, I trained it with 3 formats, square, the typical portrait and a more elongated one, all of them are trained to create designs in the center of the image, which is good because you can let your imagination fly without filling yourself with concerns like (that it literally won't work as a clothing design), then enjoy it with confidence.
PS: adding "simple background" "white background" "black background" can make it easier when you want to delete the backgrounds to print, anyway... for now I say goodbye, I have to go back to work on a lora in which I have been correcting images for days and I'm not sure how it will turn out!, bye ~ (o´▽`o)ﾉ
Description
FAQ
Comments (9)
Thank you for all you hard work and effort into making this model
Thank you for using my lora and supporting me. :3
this model is totally broken. I tried to use it and just got blank squares of color :(
Hello, I honestly don't know why the lora failed you. I usually generate things in tensor and locally so I didn't know about this error, I just tried it in the civitai generator and I recommend animagine, I don't remember which model I trained it on, I'm sorry for the bad experience.
yeah if I use the promp from the cover it works, but it fails with my own promps. even if I do something simple like "cat" I do not know why. Makes no sense
@WildRabbit I'll see if I can train a version with another labeling, do you have a specific model? illus, xl, pony, etc?
@Herrscher_AGGA no worries. ^^ if it does not work its probably something I am doing wrong since the original promp works fine. but ty for the offer
Through experimenting with this lora, I have noticed it doesn't like a lot of Checkpoints that are trained on realism/realistic. The type of model I suggest you trying this on a model with more illustration type trainings. Overall this Lora is beautiful. Thank you creator.
@ZUSIMO Thanks, it was actually one of my first SDXL parrots so I used the SDXL 1.0 checkpoint, but since there was no realism in the examples it limited its use a bit. There aren't many t-shirts with human faces like this style.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.