First time creating a proper merge. The purpose of the model is to get stability of WAI and style variation of NTR Mix to be able to create pictures that doesn't have default "AI style".
Prompting:
[your prompt] [quality tags] [artist names]Prompting is the same as Illustrious based models - booru tags with some natural language.
To get insteresting styles combine multiple names of various digital artists. Model seems to recognize artists with more than hundred arts.
I prefer to mimimize the use of negative prompt, because it makes images more generic.
Quality tags:
My personal choice, can be anything that works with Illustrious models:
masterpiece, best quality, newest, absurdres, highres, newParameters:
CFG: 2~4
Steps: 20-30
Samplers: eulers, any of DPM++
Scheduler: karras, simple, uniform
Description
(1) = 0.7(ntrMIXIllustriousXL_xiii) + 0.3(Illustrious_waiNSFWIllustrious_v120)
(2) = 0.7(Illustrious_waiNSFWIllustrious_v120) + 0.3(novaAnimeXL_ilV80)
NWNCA_Mix = 0.7(3)+0.3duchaitenCoolAge_v30
Shortly: NtrMix Xiii + WAI NSFW v12 + NovaAnimeXL v8 + DuchaitenCoolAge v3