Ani10 🎨 [Experiment]
--[Pinboard Updated🥝]--
This LoRA is trained on a specific style the dominated the anime scene from 2008 to 2010 think stuff like (Rosario to Vampire, Shakugan no Shana,early High School DXD ...)
I'm really fond of that style and i wanted to have it at hand hence this little experiment ..
The result was more general then initially intended , as it picked up the concept of anime scene , but hey I still like how it turned out .
Usage Tips (normal) :
I personally use it at the weight of 1 so i recommend that
the LoRA needs to have its trigger words "
nv-ani10, 2010" in the promptreduce the quality tags you're using, just keep "
masterpiece" that more then enough as the LoRA is trained to handle it without breaking the style.the LoRA only tested on WAI-NSFW-illustrious-SDXL V14 (used in the showcase pics)
Usage Tips (advanced)✨ :
just recently tested [NAI-XL vpred1.0 2d accelerated] and it works like a charm with this LoRA
use with the sampler "Euler Ancestral CFG++" and a "CFG = 1.6"
activate GFG ReScale and set its value to 0.5
About Training :
the dataset included I wide range of:
expressions
fantasy / scifi / modern setups
SFW and NSFW data
male and female characters
all while trying to achieve balanced distribution ration ratio
the dataset was cropped manually into sdxl native resolutions for better training fidelity while taking care not to remove essential parts
the LoRA was trained just enough to get a good effect for its dim (16) , while still having high flexibility
then some of the LoRA blocks were later tuned based on tests and personal preference
Usage Tricks :
It seems this LoRA have picked up on concept of tv anime styling , so you can nudge it out of if original dataset if you set the weight to 0.7 and add the year you wanna explore for example 2006 and it might just pull it off ! here are the plot tests for that
XYZ Plot tests with the LoRA on/off
XYZ Plot tests with the LoRA on/off +"anime screen shot" added
more tests can be found in this post
