V3
Did my best to clean up some stuff, tried to make it a bit more flexible and a bit more cleaner.
Had to actually fine tune a model a bit to get the results I wanted, then I merged a crap ton with some customs.
Overall I tried my best to get it similar to V1 & V2 with cleaner line work and more flexibility
♥ Be sure to post your creations, looking forward to seeing them!!! ♥
Sampler: Euler or Euler a
CFG: 3 - 5.5
Steps: 25 - 30
Please use ADetailer, Denoising Strength to 33 - 35 [Don't forget to toggle it ON]
Quality tags I like to use: (masterpiece, best quality, absurdres, highres), amazing quality, very aesthetic, volumetric lighting, digital art style, drawn aesthetic, concept painting, ultrahigh definition or masterpiece, best quality, absurdres, highres, extremely high detail, digital art style, drawn aesthetic, concept painting
but you can use your own or none, after testing this model more I realized that quality tags matter as well as what you put in negatives.
Negatives I like to use: worst quality, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form, feral, semi-anthro, boring, ugly or lowres, (worst quality, bad quality:1.2), bad anatomy, sketch, jpeg artifacts, signature, watermark, oldest, censored, bar_censor, simple background, conjoined, against wall, bad hands, mutated hands
but again you can use what you want
WARNING!!
Be mindful of using LoRas on this model, It doesn't need a lot of LoRas to kick ass.
One of my biggest goals was to create a model that didn't need 50 million loRas to look decent, it can work with loRas but be MINDFUL of what you are using and stacking.
The model does listen to a bit of artstyles.
V2
This model is a carefully crafted merge of custom models I’ve put together. It’s one of my more refined and “tamed” creations, though I might upload some of the wilder, more chaotic versions if there’s interest.
With this version, I aimed to improve on V1. While it takes a bit more effort to prompt, it responds well and delivers solid results once dialed in.
Let me know your thoughts, and I’m open to feedback or suggestions for future updates!
CFG: 4 - 5
Steps: 20 - 30
Can use quality tags or without.
Be sure to post your creations, looking forward to seeing them : )
Started messing around with Illustrious and decided to make my own model that fits my tastes
Keep in mind I am still getting used to Illustrious and NoobXL so this is experimental and beta.
Hands are a mess and I plan on fixing that issue hopefully in future models.
This uses danbooru tags for prompting
Just wanted to upload it and share, incase you guys wanted to use it
Be sure to upload your creations, I would love to see your work.
Please use Euler (not euler a)
Steps: 25-30
CFG: 3.5-6
face fix is a must 😬