This is my first attempt at creating a realistic model in Illustrious. After receiving a master class from QH96 on how it works and the differences with Pony, I decided to venture into Illustrious. I immediately began to notice the differences, mainly in understanding the prompt and many qualities and I fell in love immediately, I spent countless hours trying to perfect a model that would adapt to my tastes with what I had experienced in Pony and finally arrived at a result that I charm and I found it worth sharing, you will surely find some flaws since I declare myself not an expert but I hope you like it as much as I do, it is important to follow the parameters since it has many limitations, especially in terms of samplers, the parameters that I used are :
The VAE is not baked, use SDXL_vae
Upscaler: Im using Latent(antialiased) upscale 1.5 Steps 6 denois 0.35 Hires CFG 1.2, sometimes i use Remacri_original, less realistic but with cleaner images
Width and height: 896*1152
CFG Scale: 1.2
Steps: 8
The sampling method i use is Euler A-SGM Uniform
I don't like to put my models in early access because I like everyone to have access from the moment I publish it but any tips and Buzz are welcome.
Description
FAQ
Comments (11)
Amazing model. I finally feel ready to explore IL models thanks to your work here!
can you explain in few words what are the main differences between this and Pony models ? I missed the buss I guess, went off civitai for a few weeks, when I came back almost all models that post up are IL models, and I don't really know what they are :) Except they seems SDXL ?
PONY happened when someone retrained SDXL so extensively that it became basically its own thing. Same thing with Illustrious, which is another retrained SDXL, but one year later and with hopefully more improvements. Generally, Illustrious is supposed to have better anatomy, better prompt adhesion, and know more characters and art styles. It was also trained on the danbooru tags instead of whatever PONY was trained on, so you may need to rewrite prompts.
@xegoteb757932 oh, okay, that's why on the few attempts I tested, images weren't looking like anything worth posting :) guess I need to see what those danbooru tags are
@xegoteb757932 What the hell? Pony captioned exatly the booru tags, what are you smoking?
Ohh so you went straight to realism huh, awesome, the model seems pretty good, curious about the low steps and cfg tho, but the only problem I'm seeing so far is some repetitive faces and slighly problems to follow the prompt because of the low cfg I guess,overall an epic start!
This model likes to add logos/text even when heavily negative prompted not to.
Could you please share the version without using the accelerated(like hyper or lighting or LCM) model?
which one is more prompt obedient???
pony or IL.???
IL
It seems overtrained and very difficult to prompt anything fantastic or unusual. Very simple stuff comes out extremely realistic though.
Simple prompts for fun: "man", "male", "man in star wars", "a woman dressed as a man"