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
Higher contrast
Higher details within backgrounds
NoobAI EPS v1.1 + Illustrious v2.0-stable + ChenkinNoob v0.5 DARE applied
FAQ
Comments (18)
Thanks for v19. I love it
wow,ill try this
you MUST also try NovaFlat by this author!!!
This is the BEST model I ever tried, anime is greate indeed, but Flat is just better much!!
Howerver I havn't seen that update these day which is a little pity...
*Howerver I havn't seen that update these day which is a little pity...*
Is it possible to improve on perfection? (с)
@Twostep Yes, It can learn more new info, character, or try some design!
@kaedamanri730 Like many other authors, I almost never use the latest version if there are more than three of them. Authors almost always change the style or even compromise the flexibility of the models. It’s not normal to have 9 versions of a single model—and it’s even more abnormal to have 15+ versions of a single model.
Why don't I get the Nova Ex-Anime model?
Images from v19 have this weird "texture", like someone put a semi-transparent layer of bubbly colored noise over the image. It's especially visible when comparing with an image generated from v19 under the same seed, since the composition is similar.
Is there something I should be doing fundamentally differently when using 19 compared to 18?
EDIT: Flat V9 seems to have similar issue, though it's slightly less visible there.
Ok, weirdly this seems to be a sampler issue.
I normally use DPM++ 2M (it seems to result in a bit more complex image? plus Euler a needs 20 steps to not have artifacts in eyes, while this does well with 15) but both for it and Euler (non-a) the above issue happens. It doesn't happen for Euler a (which however produces significantly different image, and imo worse aside from the mentioned issue).
But weirdly this issue didn't happen on previous models (I think it might very vaguely appear on 18 but seems to not be present on 17, those are the only two I have right now, but also never noticed it with older ones and I always used DPM++ 2M).
how many upscale steps do you recommend and what upscaler works the best?
Pls bring back the greatness of IL V9-V11 😇
Any way to get rid of that AI feel/sheen? NoobAI handled it perfectly, but their dataset and limb positioning is awful.
v17 still seems to be the best. v19 always wants to give me nudity, no matter how hard I try to not get nudity.
Hey! It's an amazing checkpoint, could i use it on TensorHub?
V18 is better than v19. Why should it be disabled? V19 has issues with the prompt "90s anime/retro" and the results are choppy/unsmooth. While v19's advantage is its responsiveness to prompts, its only downside is its choppy output.
I feel version 18 is better because it allows for lighter prompting. Version 19 has overly saturated colors, making it quite difficult to control. Especially when using prompts like '1990s \(style\)', the results feel a bit off. That being said, the poses are excellent.
Details
Files
novaAnimeXL_ilV190.safetensors
Mirrors
novaAnimeXL_ilV190.safetensors
NovaAnimeILV190.safetensors
novaAnimeXL_ilV190.safetensors
novaAnimeXL_ilV190.safetensors
novaAnimeXL_ilV190.safetensors
novaAnimeXL_ilV190.safetensors
nova-anime-xl.safetensors
novaAnimeXL.safetensors
novaAnimeXL.safetensors
nova-anime-xl.safetensors
n0U4_x1x.safetensors
novaAnimeXL_ilV190.safetensors
novaAnimeXL_V190-Illustrious.safetensors
novaAnimeXL_V190-Illustrious.safetensors

