You can join my discord server here!
You can also find me on Patreon and Ko-Fi!
A weight of ~1.0 is recommended (You might want to adjust the weight depending on your model).
Feel free to experiment with other weights and share your results if you want to!
Please use the following prompts:
Base appearance (for custom outfits):
Haj1me, purple eyes, multicolored hair, grey hair, purple hair, short hair, two side up, hairclipDefault Outfit:
Haj1meDef, purple eyes, multicolored hair, grey hair, purple hair, short hair, two side up, hairclip, white choker, white shirt, white jacket, gradient jacket, white shortsReGLOSS Outfit:
Haj1meReglss, purple eyes, multicolored hair, grey hair, purple hair, short hair, two side up, white beret, hat ornament, shrug \(clothing\), shoulder cutout, white leotard, black belt, belt pouch, white skirt, white shorts, shorts under skirt, thigh strap, white thighhigh, single thighhighLet me know if you have any issues.
Description
FAQ
Comments (4)
I am sorry for asking this here. Can you tell me what is the difference b/w illustrious and noobai checkpoints? I searched on the internet but couldn't find anything for a non technical person like me.
Can I use illustrious loras for Noob and vice-versa?
I use sdxl controlnet, will it also work with noob?
@JHJH369
NoobAI is a more finetuned version of Illustrious.
NoobAI models have more knowledge and can do concepts that Illustrious does not understand out of the box.
Important here is that NoobAI models are split into two major groups 1. Epsilon Pred and 2. V-Pred.
V-Pred is more difficult to use and very annoying to train on which is why I wouldn't recommend it.
My Loras are trained on NoobAI Epsilon pred 1.1.
To a certain degree you can use noobai epsilon loras on illustrious and vice-versa, but there will be a loss of quality. Outfits will be less accurate etc. You will get the best results using noobai loras on noobai ckpts, but nothing stops you from trying noobai loras on illu ckpts or the other way around. It just might yield you worse results than you might expect.
NoobAI has their own set of controlnet models which you can find here.
Hope this helps
@ChameleonAI Thank you so much. It is somewhat like, what illustrious was to pony, the only difference is that it is a further enhancement of illustrious.
I needed to know this because I have limited internet data to download models and test on my own.
@JHJH369 Pony and Illustrious are more different from each other than Illustrious and NoobAI, but yeah kinda. It's still all roughly the same architecture
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.