UPDATE (July 10th 2026):
I did it, Finally finished PawPaw Primal! only took me way too long, putting in early access, because I had to do some Lora Training for this merge... So I wanna try getting some of the buzz back...
Model Information:
A Model made as the successor to my SD1.5 model line called Furation, which was designed to mimic the same style of SD1.5 with everything from SDXL, PonyXL, Illustrious and NoobAI that makes it better than SD1.5. it's meant to have a 2.5D style by default but it can do many styles, I just personally love the 2.5D style. Please don't use the SDXL version because it sucks, but the pony and all of the Noobai/Illustrious models are really good
Prompting:
Noob:
Positive:
masterpiece, best quality, amazing quality, very aesthetic, high resolution, ultra-detailed, absurdres, newest, scenery, furry, anthro, detailed background, detailed hair, depth of field, detailed fluffy fur, photorealistic details, volumetric lighting, BREAKNegative:
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 digit, fewer digits, cropped, very displeasing, (worst quality, bad quality:1.2), bad anatomy, sketch, jpeg artifacts, signature, watermark, username, simple background, conjoined, bad ai-generatedPony:
This model has way better prompting so using simple prompt templates like the following:
Positive:
detailed background, detailed foreground, depth of field, soft lighting, volumetric lighting, (kemono), realistic, digital media \(artwork\), commission, absurd res, BREAK
Negative:
malformed, worst quality, bad quality, (ahegao), bangs, bad eyes, messy hair,Can produce Great results even without further prompting, But if you do want some suggestions to get even better images, Use Negative hands and Boring SDXL
Also Just use E621 Tags, you could use any Booru tags but it doesn't work as good with Booru tags..
Description
FAQ
Comments (10)
The quality of the images generated in this model is really good, but I noticed the image are being dark most of the time and there are random white spots in the image which kinda ruins the aesthetics.
Weird... Ikd what you're talking about, but it could be an issue with whatever backend/ui you're using since I use ForgeUi and I've not found any issues. Or it could be an issue with your base prompt. But if you have some examples of the problem I'd be willing to look at them and see if I can replicate or fix them.
Can’t use it, says “some resources used to generate this image is unavailable” Is it up for auction or something? I don’t know how it works (I’m new and on mobile btw)
the model hasn't been given on site generation yet sadly
@Masterjp123 Oh alright, thanks for a response
Works great with Illustrious-based loras but you gotta load separate VAE if you are having the "white dots on black" issue or the colour wash. Used it with default SDXL VAE and works perfectly for me even with a ton of loras and other stuff. Great work and I'm looking forward to see new updates!
yeah I'm planning to fix the VAE issues next release
this has been really useful for doing img to img reworks for my OC
I'm getting white washed, low contrast results. I'm using Euler A Karras at 40 steps and 4 CFG. The colors are very bland. Any ideas?
Don't use the Karras scheduler, NoobAI and Illustrious based models hate it.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.





