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
Make it to have the look of Illustrious v1.0 while having better posing
Noob EPS v1.1 and Illustrious v1.1 DARE applied
FAQ
Comments (23)
Crody's Nova models never disappoint! Thank you!
The latest versions, v5A and v5B, are somewhat disappointing compared to v3 and v4. Perhaps this is because they are based on NoobAI. Literally, all the artworks I generated on v4 are more stable and consistent in color correction than those on v5 and v6. I will continue testing v6, but at the moment, the NoobAI-based versions significantly fall behind in the key parameters I focus on when generating images. I’d love to see a new version based on Illustrious, featuring improved generation of hands, eyes, bodies, and lighting. P.S.: However, v6 might be on one level with v4. I should to use v6 such more.
Just realized, there is NO confirmation of any kind when you are writing something here and accidentally clicked off of the text box (which deleted my comment), literally zero warning and it just deletes it, awful design
Nvm, entirely new structure because of Illust V1.1, now I have to suffer to redo my setup again
still decent though
Sorry, what are you talking about and why are you doing that in a comment for a model?
I have zero idea what "bold text" and "text box" you are talking about. This is a safetensor file you just load into a workspace of your choice, with whatever text you fancy.
@Vospi Oh no, I was talking about when I was writing this comment and it just deleted my shit without any confirmation :sob:
Glad people didn't have the capacity to read ig
This is my first Illustrious checkpoint. I'm so excited!
Could you please bring back the v4 as it's way better than the v5 and v6? Appreciate that!
Works well but needs some finishing touches Works well but needs a little work to get a good picture
Pretty nice one 👍
Gotta be honest 4.0 is way better than v5 and v6.
Please upload your pics to show the differences between them.
After testing for a week, I can say that both v6 and v4 deliver really solid results. Now let’s talk differences.
v4 with Euler_A gives more detailed objects and richer colors than v6.
v6 with DPM_2M matches v4 with Euler_A in terms of detail. Honestly, it’s hard to spot the difference.
v6 with Euler_A has softer, more pastel anime-style tones and lower color saturation compared to v4 on the same sampler.
v4 with DPM_2M is way too sharp. No matter what I tried, it kept over-sharpening the objects.
You can use either model depending on your needs. Personally, I give a slight edge (about 5%) to v4 since it plays better with most LORA characters and brightness stabilizers.
If you're using HiresFix anyway, go for v6 with DPM_2M — it’ll give you great detail, good saturation, and it won’t go overboard on sharpness like v4 does.
What are the best hires settings for this model ILv6?
To me x1.5, 0.4 denoise with 40 steps (Which equals to 40*0.4=16 steps) is the best
The sample images are made with this setting
I've been very satisfied with this model so far. Handles a decent variety of styles pretty well.
absolutly good
My favorite model
Great checkpoint. Supports many artist styles
what a base, nai or ilxl?
It's both since the base is combination of Illustrious and NoobAI using DARE merge
Cute
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.