Anima Version is Available!!
Nova Anime XL
Nova Anime XL is Nova Anime: Anime/2.5D/3D checkpoint model for SDXL
Pony version aims to get the similar look as both Nova Anime and Nova Domain
All images on example were created with diffusers with custom png tags
PixAI - IL v4.0 Happy Valentine
PixAI - IL v3.0 Happy New Year
PixAI - IL v2.5 Merry Christmas
Rules
You cannot use the generated images for commercial use if it's not edited (or just turning it to black and white)
You can share images without any restriction if you don't monetize it
Advertising this model to outside is always welcome
Recommend Settings
Sampler: Euler a
Steps: 20~30
Clip Skip: 1-2
Denoising Strength: 0.65 - 0.8
(Pony)
CFG Scale: 5~7
Prompt: score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime, BREAK
Negative Prompts: score_4, score_5, 3d, jpeg artifacts, username, watermark, signature, normal quality, worst quality, large head, low quality, text, error, missing fingers, extra digits, fewer digits, bad eye
(Illustrious)
CFG Scale: 4~6
Prompt: masterpiece, best quality, amazing quality, 4k, very aesthetic, high resolution, ultra-detailed, absurdres, newest, scenery, {Prompt}, BREAK, depth of field, volumetric lighting
Negative Prompts: modern, recent, old, oldest, cartoon, graphic, text, painting, crayon, graphite, abstract, glitch, deformed, mutated, ugly, disfigured, long body, lowres, bad anatomy, bad hands, missing fingers, extra digits, fewer digits, cropped, very displeasing, (worst quality, bad quality:1.2), bad anatomy, sketch, jpeg artifacts, signature, watermark, username, signature, simple background, conjoined,bad ai-generated
Description
Changes
Much stable / dynamic posing
Stable Fingers
Enhanced visuals for male characters
Brighter / Vivid visuals
NoobAI EPS v1.1 + Illustrious v2.0-stable DARE applied
FAQ
Comments (18)
Will you be making a Anima version of your great Checkpoints? Your Checkpoints along with WAI0731 are definitely my favourite. You both put so much time and effort into making them. Well done.
Still great model till version v17.
But may I request to update or extend the recognized characters list on next update?
Probably one of the best straightforward anime finetunes. 12.5 seems to look the best for me
Definetly brought back the color/contrast with the v17.
Wonderful... Quite elegant!
v17 is amazing! I feel like it adds the variety I remember back from older versions back
可愛い
The model sometimes struggle with character LoRAs as it conflicts with its built-in version of that character. Also, it is generally difficult to override the default clothing/styles of those built-in characters.
v17 is not working at all for me.... please bring back v16 :(
V17不太好用...V16比較強
V170, has an issue with xray, cervix penetration. Where in 85% of the cases also introduces internal cumshot, which it shouldnt do unless prompted for it.
Why in gods name does this model love doors
Why i get 1 extra heads? :D
waiting v18
IDK if this true for this checkpoint only or not. When using V17
It seems I get some "bake-in" result when I'm using broad tags and not controlling it whenever I'm generating using my own LoRA. Some of the cases are:
1. Dark-skinned female + nudity will give her tanlines even though I'm not prompting tanlines and there's no tanlines in the dataset
2. Shirt are often get print shirt detail even though I just prompting crop top or t-shirt, and my LoRA dataset does not have that detail either.
3. Shirt are often wet on cleavage area if not controlled correctly
4. crop top overhang are tend to generating additional bikini under shirt (the setting are in beach)
I only started using this model, but i certainly had that with other models. Putting tan lines in the negative and what not to try counter it. I also had the annoying thing where not only it put something on a shirt, but literally a similar detail level face print on it as the one wearing it. Most recent frustration were it put hanging pictures in the background with all kinds of faces on it. So these things do happen elsewhere too for sure.
@Kalderas Ahh I see.., thanks for letting me know!
V17 isn't capturing the LoRA's style very well, but V13 reflects it much better. Feels like it's starting to overfit, haha