Nova Animal XL
A realistic furry checkpoint model that can have great details on any type of furs, scales and feathers, which aims to be XL-lized Persika Furry
Rules
You cannot use the generated images for commercial use if it's not edited (or just turning it to black and white)
Advertising is always welcome
Recommended settings
Sampler: Euler a
Steps: 20-40
(Pony)
CFG Scale: 6-7
Prompt: score_9, score_8_up, score_7_up, BREAK, furry, anthro, detailed face eyes and fur, {Animal Type}
Negative Prompt: score_6,score_5,score_4, source_anime,source_cartoon, 2d, art, cartoon, blurry, 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: 3-6
Prompt: masterpiece, best quality, amazing quality, very aesthetic, high resolution, ultra-detailed, absurdres, scenery, photo, realistic, furry, anthro, {Prompt}, BREAK, depth of field, detailed fluffy fur, photorealistic details
Negative Prompt: smooth skin, human, multiple tails, 2d, art, 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, very displeasing, (worst quality, bad quality:1.2), bad anatomy, sketch, jpeg artifacts, signature, watermark, username, simple background, conjoined, bad ai-generated
Description
Improved facial structure and posing, remove background crispiness / over detailing
Noob EPS 1.1 + Illustrious v1.1 DARE applied
FAQ
Comments (13)
IL v3 is so much more usable with the noise. The best IL model so far (*might be better with V6a merge)
Well, I had great hopes for IL V3, I thought I would go to heaven, but here I am sitting and crying馃樋
The fur has become noticeably flat, the image has become larger in 2.5D than in realism/realistic, the main object has become brighter. Now we take this bright 2.5D object and a realistic background, and what do we get? We don't get a whole image, we get an object that doesn't fit into the background. The object and background are poorly synchronized. I'm disappointed.
Yes, the model has become smarter, smarter, I don't know how much, but I don't care馃挃
I'm grateful to the developer for the work, but I'm staying on IL V2
Maybe I just don't understand something, what am I doing wrong?馃槗
V2 had an issue with background and prompt adherence
To fix it, I simply decreased the noise factor using finetune and merged NovaFurry IL v6B
Because of that the fur looks slightly flat but it can improve by using detailed fluffy fur and not using volumetric lighting
If the object gets floaty, try using depth of field, photorealistic and scenery
@Crody聽Have you considered merging it with 6A instead? It gives the more "realistic" look
@furproxy聽I'll do it on next version
@Crody聽To add on to MiraKok's post, it just seems realism in general is heavily toned down in V3, definitely more 2.5D.
Incredible model, as always! Thank you for making it!
I really enjoy this model, it's so clean and so well done. hope to see more from you
Downloaded this for animals, but found out that it also generates normal human girls with amazingly good eyes, teeth and skin. A lot better than some other Nova models, which garble eyes, mouths or skin. Guess this one just takes advantage of detailed parts from animals. So this model is going to be my favorite for all cases now.
Btw.. this model (IL v3) is insanely coherent with tiled upscales (not as much with loras). Highly suggest using it for that.
IL V3 seems to be noticably less realistic, at least for human characters. I've been primarily using V2 because it outperforms even human focused realism models and this blender checkpoint i really like, especially in character accuracy consistency.