This is my first lora that I train, don't forget to support the artist https://x.com/AdamA39062
Description
FAQ
Comments (15)
Looks neat.
Also, up for art style suggestions?
Sure!
Thanks for the sexy art style! I really like the body type and the brushstroke effect. The eyelashes aren't really my thing, though.
Is it possible to remove them through prompting?
Yeah I've tried to fix it but it's still there. the best way for me is add (short eyelashes:1.45) and reduce the strength to 0.5 - 0.75. Hope that help!
@mithdam191 Alright! I will test and post images in the gallery here!
If you like this style, I'd love to see a LoRA made for it as well. This style looks amazing! 😁💗
I think it's AI-generated, possibly using NovelAI, but I couldn't find the original creator. If you could make something similar—or even recreate this style—it would be incredible.
@ikiru99percent That's look amazing! I might make it thanks for suggestion💗
@mithdam191 Thank you!
A suggestions:
It would be helpful to mention in the description that the Anima Aesthetic model should be used. It seems to be the only model that gives good results.
I tried using the base model, but the results were much worse.
I also dragged one of your images into ComfyUI to check the metadata. I saw several LoRAs, so I tried adding a few of them, but I still couldn't recreate the brushstroke effect in your images.
One question, did you upscale the images?
I didn't use any upscale bro but my checkpoint is SAMANIMA and tons of lora so I don't really know what It look like with base model
@mithdam191 Thanks for the info!
Which base model did you use to train this lora? SAMANIMA?
@ikiru99percent I used the base model to train, I tried to used SAMANIMA but it's wouldn't start so I just used the base version
@mithdam191 Ok, I will test it out with SAMANIMA ckp
@mithdam191 Sam ANIMA - Turbo v1.0 | Anima Checkpoint | Civitai
this model?
@ikiru99percent https://civitai.red/models/2409949/sam-anima-realistic?modelVersionId=3060073 this one 2.2.5
@mithdam191 It's giving me more realistic-looking outputs with pretty bad anatomy.
I'll stick with the Aesthetic model for now. Anyway, it's a great style. I hope to see more LoRAs like this in the future! 👍