Nova Kemono XL
A 2d/2.5d furry checkpoint model with eastern look
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 for best performance
Steps: 20-50 steps
CFG Scale: 3-5
Prompt: masterpiece, best quality, amazing quality, 4k, very aesthetic, high resolution, ultra-detailed, absurdres, newest, scenery, furry, anthro, {Prompt}, BREAK, detailed background, detailed hair, depth of field, detailed fluffy fur, photorealistic details, volumetric lighting
Negative Prompt: human, multiple tails, modern, recent, old, oldest, graphic, cartoon, 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, simple background, conjoined, bad ai-generated
Description
Created based on v4.0
NoobAI EPS v1.1 + Illustrious v2.0-stable DARE applied
FAQ
Comments (6)
v4 is still the best of all, both in terms of style and the accuracy of reproducing the characters' appearance.
I like V3 and V6 because this models looks like clear merges without a lot loras/overtraining/something, so with this versions i can control results what i want. And thanks Crody for all cool stuff you made!
6.0B needs to fix its artifacting problem, I basically have to upscale every image to get rid of the artifacting
I've noticed the eyes sometimes get bright green (or different saturated color) no matter if I describe the eye color in the prompt or not.
Adding "glowing eyes, green eyes" to negative doesn't fix it (eyes still get somewhat distorted and miscolored).
also some random rainbow artifacts when using artist tags (probably LORAs too, haven't tested)
@Sluggish_Python if you upscale it, it'll fix it. But that shouldn't be mandatory
It is not bad, but have some issues. Tried to make an Flareon without any loras, but a lot of things when wrong.
- It refused to make a proper front view, always slightly above.
- Neck was long no matter what i tried.
- Refused to stop making the character leaning forward, even with prompt.
- feet was out of frame no matter what i did.
- Colors was always off, weird color blobs as well.
NOTE: it was 60 B
I checked my settings, was no problems, was following the instructions, so i tested a lora for kemono style instead with same prompt to see if it was the prompts, and the result was much better
Thanks for taking this feedback, hope to see a new version soon!
UPDATE: Tested the normal 60 and it works a lot better
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.