Moe Fussion XL
Moe Fussion focuses on providing wide spectrum of models suitable for different purposes. You are free to use these models however you want to what SDXL License allows. Please read a rest to know how to use these models.

General Info:
These are ZTSNR + V-Pred models, so you will need SwarmUI or ComfyUI to run them, while they should get autodetected it's suggested to manually specify the model as zsnr+vpred.
Cfg: around 5
Sampler: normal Euler is suggested
Sheduler: normal or AYS
Steps: 30
Positive prompt for NIASR isn't really required, for NI/NIA best quality is fine.
Negative prompt: lowres, worst quality, low quality, old, early, bad anatomy, bad hands, 4koma, comic, greyscale, censored, jpeg artifacts, blurry
Prompting style: use danbooru tags for the most part
These are not absolute rules how to use the model, feel free to experiment, for some specific scenarios different configuration may be better.

These versions should contain everything that anime sdxl models currently offer while being less scuffed than most of other merges. By merging you can't really create a better model than it's best component, you can only create something which is better at something and worse with something else, that's why so many versions and these versions should be considered equal, however they are meant for different purposes.

Version Descriptions:
NI_Lx0 - is like, do stuff yourself, so you have the most control over style and stuff here

NIS_Lx0 - is more autopilot, has some soft default style

NIA_Lx1 - closer to NI_Lx0 but tries to improve demography prompts

NIAS_Lx1 - NIS + NIA

NIASR_Lx2 - full autopilot with strong default style, prompting styles will result in variations at most

I am unsure whatever I will use Civitai in future so you may follow me here https://huggingface.co/LoliRimuru/moeFussion