Trained on AnyLora
Please note that with some enhancers(Detail Tweaker, e.t.c.) it is very easy to get better results than showcased, since i didn't use any to generate showcase images for this LoRA
To get the best results, please read the description below:
Recommended weight for V1: 0.5-0.8
Recommended weight for V2: 0.7-0.9
Activation word for V1: Perona
Activation word for V2: PeronaChan
Use of " beautiful detailed eyes, " and "Detail Tweaker LoRA" enhancer LoRA's is also recommended for good results
Upscale to 1.5
Using anime SD models is recommended, but some semi-realistic models also work
I would appreciate any reviews, positive or negative!
I would very much appreciate you sharing your results ^_^
If you have any suggestions on what LoRA i should do next, please share them in the comments
Enjoy :)
Description
Version 2 - trained on 80 images
Impoved dataset quality
Increased LoRA training resolution (512->1024)
FAQ
Comments (5)
Edit: Nvmd, keep doing what you're doing, best I've seen.
These are really good and capture the likeness well. One thing worth noting is that you might not want to detail their physical appearance "Pink hair, black eyes, twintails, etc" so that they'll automatically be associated with the character. Unless they have multiple different hairstyles etc, and you want to be able to change them.
But otherwise I'm checking out some of the others right now and they look great will leave review.
Thank you for the feedback and advice! I got a habit of detailing character’s appearance for some reason, perhaps that is not a very good one. I’ll definitely check out some of yours too! The showcases look very cool, even though i’m not familiar with all the characters :)
@12user34kn276 Haha you don't have to. I've just been following for a while, but hadn't had a chance to test out the models yet. Been watching OP for years though so love the characters!
Good LoRa, it is a little overtrained but that kinda helps it nail her eyes more than other Perona LoRas. Keep it up man, we need more good LoRa trainers, especially for the one piece characters lol.
Thanks! I feel like a lot of my LoRA’s may be a little overtrained, but i personally like it that way:)