This LoRA tries to improve the outlines performance, most of the popular anime models (e.g. Anything V4.5, Counter Feit, AbyssOrangeMix, etc.) have some sort of pastel-like feeling and rather thin outlines.
This LoRA is trained on stylized images and then merged with several LoRA models I trained previously (most of them are already published on Civitai, you can check my homepage).
The example images are generated with Anything V4.5 + orange mix VAE, but this LoRA is capable of multiple models (also including NSFW), along with different character LoRA.
For strength, I'd say 1.0 is good enough.
I've also attached a checkpoint version, merged with Anything V4.5. It's for more sophisticated use (e.g. using another style LoRA).
中文：
本LoRA模型是为解决多数动漫风格模型过于柔和导致描线较弱的问题而训练的。基于风格化的图片训练，并与我此前训练的多个LoRA模型进行了融合而成。
示例图片为Anything V4.5 + orange mix VAE生成，但本LoRA并不限定基础模型的种类（包括NSFW类），并可兼容多数人物LoRA。
推荐强度为1.0。
附带的Checkpoint版本是和Anything V4.5融合的版本，用于一些更复杂的场景（如使用其他的画风LoRA）。
Description
Base model: Anything V4.5
Recommend VAE: orangemix or kl-f8-anime2
FAQ
Comments (13)
Can't wait to use this LoRA! You wouldn't happen to have a link to the Easy Negative embedding as well would you? I can't seem to find it.
It's here
This is exactly what I've been wanting, thank you
This is by far my favorite Lora. Any chance we will be allowed to use this commercially in our games ?
Yes, you can, but you may need a citation.
@CyberAIchemist awesome, for example let’s say I’m making a card game, I just put @ cyberalchemist + name of Lora in Artist Credit for the specific card where your Lora was used to generate the art ? I may have to format things differently since most of the art will be ai generated ..but in general something like this would be fine ?
@kuzenghazel378 Yes
Can you upload a non-safetensor version? I am having issues trying to load it with the hugging face diffuser. I just realized this isn't a LoRA but a LyCORIS. Can you update this?
Hi, is it possible to make opposite LORA. Thinner Lines Anime Style:) I get excellent result with model+manga like lineart lora, but I get very thick lines. It would be great to make them thinner and connected like on this LORA.
Quick question, when you say the checkpoint was merged with anything v4.5 do you mean this one;
https://civitai.com/models/4855/anythingelse-v4
Thanks for the amazing LoRA and model, I use it everyday
Very interesting fact about this is that you can use it with negative weight to make illustrations (photo)realistic without changing checkpoints or anything of the sort.
Good point - chers! I'm enjoying using it to switch pics back and fortha at the mo!
This lora produces beautiful eyes. :)