STICKERS STYLE work 1 weight enjoy it.Lora xl with dpo technology experimental natural language:
Experiment with the scores to get the desired details with the promt:
score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up,
READ: works with specific models with dpo technology, do not try them with other xl models because it will not work:
Best models:
Pony Diffusion XL V6
AutismMix SDX
triggers words: stickers, outline,transparent background
Comisions on Fiverr Patreon Twitter
Description
FAQ
Comments (13)
seems your prompts cant make imgs like this with the lora
edit: my fault, it need some conditions to work
you tried I'm going to do the test you put weight 0.6 and you placed stickers
@Creativehotia Yes I tried 30 more even with exactly your prompts, with 10 more models, no one like a sticker, I guess I can wait for the others to see what they get.
@dajusha i had the same result, the better model for this Lora in my opinion, its RevAnimate
<lora:stickers:0.6>((stickers)) anime high quality it works -.-
@Creativehotia @YeiYeiArt Yes, RevAnimate v11 worked, in some conditions. I made 2 tests, please see my post.
wtf 8k in prompt really is anime no realistic
Hi, do u have any recommendations for dimensions? I'm using the rev animated model, sample prompts, sticker lora and have only managed to get 1 true sticker. the rest have been full images or images with a artist desk in the background.
You are placing the indications correctly I use 768x768 you downloaded the lora correctly
Well, I tried this lora and it couldn't draw any sticker with a little more prompts, but,
This lora can 2d the picture quickly and add some cute accessories, inpaint is ok, enjoy~
leave some very clear indications of how to use something they are doing wrong lora 2d is one of the most popular in my library this is a sticker
@Creativehotia It doesn't matter, I ended up using other Lora, and compared to them, the probability of generating stickers with this one is really low. though I really like the printmaking-like style of your preview images. Just for your reference.
Probably prompts more than 20 it won't work well, even with high weights. So it's hard to generate something with specified contents, except for random. I hope it will be improved.