Styles are trained on synthetic data and do not copy the styles of artists!
OIS (One Image Style) - These LoRAs are trained on only one image, these styles are just proof of concept. but the style is as stable and consistent as possible.
I do not guarantee that these LoRAs will work on other models based on Illustrious except NoobAI-XL
Recommended LoRA Strength (Weight): 0.6-0.7
Preview generated without using ADetailer & Hires. fix
Description
FAQ
Comments (3)
Have you tried WAI NSFW v7/V8, its a very refined illustrious checkpoint that not only provides consistent output but is also very stable but that's just my opinion, i wanna know your thoughts on it
Yes, I tried this model. WAI-NSFW-illustrious-SDXL v7.0/v8.0
I don't even know what to answer, it's a good model and many people like this model
This model has good compatibility with LoRA trained on NoobAI-XL and Illustrious-XL, as much as possible.
I haven't tested this model much, but it works well.
@prgfrg23 don't you think its more stable than noobaixl and such? in my testing i have found out i get more consistent results with WAI, its crazy good but idk why its not as popular as noobaixl and its different variants. Lot of resource creators on civitai were dismissing illustrious as just hype probably because of its inconsistent results but after i told them about WAI, they've completely changed their minds. Honestly i am just tired of the billion different noobaixl variants so i just stick with the simple checkpoint