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
Reconfigure overall details based on v12.0
NoobAI EPS v1.1 + Illustrious v2.0-stable DARE applied
FAQ
Comments (31)
I feel like 13.5 is a downgrade in terms of flexibility. Underwater images, for example, only have like 5 poses now. Also, there is a Kion lora I like to use and the model keeps messing up his face by making it look like its season 3 version instead of seasons 1 and 2. Despite this, I noticed that it's more consistent. If v14 has the flexibility of V12 and the consistency of V13.5, it'll be an amazing model.
Why do I get this error? I'm using Easy Diffusion 3.0. I haven't installed anything extra. I don't understand what to do. Error: Could not load the stable-diffusion model! Reason: 'Embedding' object has no attribute 'wrapped'. V12 doesn't work for me, but V10 works.
this model uses danbooru, e621 or gelbooru tags?
I jumped straight from 12 to 13.5 and let's just say it refuses to weigh no matter how many parentheses you throw at it. Stay with 12 folks! Its gens are absolutely gorgeous.
Hey, nice model, I like to use it a lot!
But I encountered an issue (probably on my end) when I want to set the fur color of a character. Do you have any advice on how to prompt that? Tried to search the tags in the listed guide you shared here but couldn't find anything useful...
Try using color + fur tags on e621 (eg. blue fur)
@Crody Thanks, I'll try it again. Although sometimes the colors are switched or present only locally, but I think that's a prompt issue on my end.
Coming from 12 to 13.5, I am wowed how much the quality jump is and how much instead of being heavy handed with my body shaping prompts to get normal proportions with some character LoRAs and constantly requiring eye enhancer.
13.5 delivers for setup I have dropped the eye enhancer and my heavy handedness for body shaping prompts is better rewarded. I can finally escape models default to thick bodies!
Personal preference but aside from the lack of flexibility of the V13 models, it's been the one I use most lately because of how nice and soft the colors are. I don't really like how punchy and saturated V12 is, so I think the slight trade for calmer colors with flexibility is fine for me. Hopefully something with more flexibility and the same colors/lightings as V13 comes out!
I was wondering if this model can be used to create photorealistic images. I'm trying to generate realistic images of characters like Peach or Nami in real life, but I'm having trouble. Do you have any tips or suggestions?
@raunen0 Try adding Illustrious Realism LoRAs.
You can also use Nova animal instead, it has great results and doesn't require a LoRA
@lnomsim Yes that can work too, but it's more-so meant for anthropomorphic creatures, not realistic human/human-like people like poster mentioned. If not, try Nova Orange XL with Illustrious Realism LoRAs cranked up.
considering they are asking about THIS MODEL which is a furry model, I thought they were asking about realistic furry model.
I'm so unbelievably slow that I forgot to give reviews for like the last 3 versions or something...
Out of those three version you are reffering too. V12 was the best, then V13 and V13.5 had some issues, but V14 is gold.
Problems with Comfy:
It works great for me on SB/Forge, but on Comfy all the images are blurry/smudged. Neither Vae nor Clip Layer help.
EDIT: It turned out to be simpler—just work with this model in EULER_A.
@Twostep Try this workflow: https://civitai.com/articles/23018
@123inactace The whole problem was that I didn't use EULER_A.
*V13/V13.5 tips that are working for me.*
For anyone using DMD 2pass workflows in comfy : LCM Karras has been great at the first pass for me.
(my comfyui flow is 1st pass > Upscale > 2nd pass > Detailer)
Also, If you're lacking some vibrancy after second pass, i recommend an image adjustment node that has auto color/saturation settings like this one
if im being dumb and y'all got a better technique, pls let me know <3
I have a quick license question:
If I draw the main lineart myself (even very simple sketch), then use ControlNet Lineart + Nova Furry XL to generate the final image, can I use the resulting images commercially on Patreon (selling high-res, PSDs, etc.) without further editing?
This workflow would mean a human created the primary lineart. Would this be considered “significant human creativity” under the Illustrious License?
Thank you in advance!
Yes, you can sell it if that's the case
Ver. 12.0 its just perfect, but ver. 13.5 for some reason is unable to do flatcolor and continues to do it "semi-realistic" without many changes for some reason :/
I've been using these models since V5, and I've got to say that 13.5 nailed it!
The prompt understanding is superb, the color pallets are great, and overall, the characters look so good.
Great job! Thank you so much!
I don't know whats wrong now, but I only get an error message, when trying to run the model. Did I miss anything to set up?
Error: Could not load the stable-diffusion model! Reason: 'Embedding' object has no attribute 'wrapped'
Currently working on the fixes
It's known issue with 6.65G Nova models
The next version won't have the issue
@Crody Okay, thanks for the info. Then I'll wait for an update.
@Crody Version 14.0 works again. Thanks for the update. Keep up the cool work.
btw. did you ever thought of making also a model for img to video? I guess that would be ultra if one had a furry model for that.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.