Trained with SD v1.5 ema
Recommended weight: 1 --> 1.1
Recommended CFG scale: 5 --> 6.5





♡⸜(˶˃ ᵕ ˂˶)⸝♡ leave a follow & like if you like what I do! much love ♡
Trained with SD v1.5 ema
Recommended weight: 1 --> 1.1
Recommended CFG scale: 5 --> 6.5
This is a LoRA model that replicates the art style of the YouTuber Kooleen! (pls and support Kooleen; she's awesome!)
The higher the weight, the more Kooleen-like it becomes, so beware!
Generating higher resolution images with high sampling steps gives superb results than using lower resolutions and lower sampling steps.
Requires no keywords such as [Kooleen Artstyle, Kooleen, etc] to function.
Use keywords such as (low detail) in positive prompt, or (shaded) in negative prompt to make the artwork more inline with the Kooleen art style.
I found using concept switch [soft image : low detail:15] provides pretty good results, so experiment and find out what works best for you!
Works great with protogenv22anime, haven't tested others, but I assume it works well with them too.
Thank you for reading, and thank you for downloading!
Trained with SD v1.5 ema
Recommended weight: 1 --> 1.1
Recommended CFG scale: 5 --> 6.5
1. 轉载模型僅供學習與交流分享,其版權及最終解释權归原作者。
2. 模型原作者如需認領模型,請通過官方渠道联系海藝AI工作人員進行認證。我們致力於保護每一位創作者的權益。 點擊去認領
