Anima version is available!
Nova Furry XL
A 2d/2.5d furry checkpoint model that can have great details on any type of furs, scales and feathers, which aims to be XL-fied Peaki 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 for best performance
Steps: 20-50 steps
(Pony)
CFG Scale: 6-7
Prompt: score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_furry, BREAK, furry, anthro, detailed face eyes and fur, {Animal Type}
Negative Prompt: 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: 3-5
Prompt: masterpiece, best quality, amazing quality, very aesthetic, high resolution, ultra-detailed, absurdres, newest, scenery, furry, anthro, {Prompt}, BREAK, depth of field, detailed fluffy fur, 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
Better posing
FAQ
Comments (19)
Really glad to see this model makes it's way towards Illustrious. I look forward to seeing it progress
Illustrious/NoobXL v1.0 releases next week!
Bro, that illustrious is absolutely fantastic. Fkn legend my man
Whenever i try inpainting the inpainted area gets a greenish tint. Any suggestions to avoid this? (AMAZING model btw!)
This might help you
@Crody thanks!
@Crody I was able to fix the issue! changing the settings settings>stable diffusion>Apply color correction to img2img results to match original colors helped a little but prompting negatively helped much more. (green tint, green hue, oversaturated:1.5), solved it entirely.
Is there a prompt combination to get this kind of art style with other illustrious models?
I don’t think there is
I merged some LoRAs to get this look
If you want to get the look with other models, you can make your own model by merging this model to it with block merging OUT00-04 with 1
eg.) Model + NovaFurry "0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,0,0,0,0,0" Merged
For those who wondering about v1.0IL:
Eyes are rounder which look similar to Pony models, details are much better and the base is same as NovaOrange v3.0: Noob1.1
My new favorite model! But I've ran into a weird issue with this model specifically where after a few gens the model will only output fully black images, sometimes after 100 images, sometimes after just 4. I switch to a different model and it works perfectly fine. If you have any insight on a potential fix I'd love that :) Either way amazing work with Illustrious!
There are 3 possible and fixable reasons for that
1. GPU RAM overload
Using too much VRAM can cause images to be worse and sometimes becomes gray
This can be prevented by using xformers, or use UI less system like diffusers + sd_embed
2. Token issue
SDXL models have special token process that sometimes the token become negative and that causes images to be worse
For example, Animagine cannot comprehend eni
There is no way to fix this so I suggest use different prompt
3. Token overload
For some reason, Illustrious/NoobXL gets worse when you use too much tokens
Just don't use too much and that'll fix the issue
I suggest using diffusers + sd_embed on kaggle
That way you can prevent any of those above and can generate images with same quality
You can create the notebook by following the first article on Suggested Resources
@Crody assuming its the token overload, is there a range you suggest to stay within?
@KD67 Around 256
Very good model. One of my favorite all-in-one out there from IL, not just for furries. The only thing is that the generations are a bit "dirty" and "messy", cloudy, I don't know how to describe it. Usually requires some additional work to clean that up. Perhaps I use a wrong sampler.
Did you use DPM sampler?
Use Euler A instead
I'd describe it as 'muddy' it reminds me of watercolors or daubed painting. That said, this is by far my favorite model currently and I usually run images through a couple of refiners and upscalers to get a sharper look overall. I've also be running with various samplers and config at 5.5. It's not a bad effect per se, it's just always there