The other day I found a embedding of DAGASI's art style but it was too bad to create some satisfactory pictures. So I made a lora of DAGASI's art style and it is not bad. The lora model is based on Novel model and the example pictures were got from grapefruit v4.1. The recommended lora weight is 0.6.
disclaimer
All my lora models are only for communication. Please don't use my models to create illegal pictures and the responsbility for generating relevant pictures will be brone by the pictures' prompt operator himself. What's more, if my lora model violated your copyright, please contact me and I will delete my model and give an apology to you.
Description
FAQ
Comments (15)
Oh my! I was doing the same thing lol
It'll be interesting to compare how they fare side by side :D
Absolutely interesting my friendXD. DAGASI is one of my favorite artists!
This is very good. Its eyes and hairstyles look very feeling, I will try ~
your crosskemono_b model is one of my favorite models ~
Oh, a man of a culture I see. Good job.
Did you train it also on his pokemon/animal crossing stuff or only "classic" furry?
I chose 98 pictures from DAGASI's vertical works to train this lora and some of the pokemon/animal crossing contents were included. But I didn't enter the characters' name in the training pictures' tag files. you could give it a try, maybe it will work :)
Does what it says. Nice.
heyo, i see that the trigger words says "art by dagasi" but does "<lora:furryArtStyleBy_v10>" work to look just as good?
The trigger words I set when training just "art by dagasi". And this trigger words are not necessary. But I didn't try the type of tag "<lora:XXX>". The lora always work when you go live it.
You know why you are here. Yes, it is good at it too.
super adorable and cultured LORA. it not perfect but it still work pretty well
When I try to use this one I get a message about a upres missing. Do I use this one like a regular model or do I need to use it in combination iwth something else somehow?
I'm sorry but actually I don't know the reason of this problem. This lora model was trained by using training script and it may not suitable for some settings. Maybe you could try using it like a normal Lora model without other settings like Hires.fix, upscaler or something.
Can you add Klonoa and King of Sorrow?
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.