Functionally speaking, this model will be worse at just about everything other than 2 specific functions: it makes thick lines about on par with other good models, and the only thing it excels at: making thick, angular eyelashes. I understand that this is confusing, so I will include a short series of questions I think I should answer - obviously feel free to ask more. You could already get those almond shaped anime eyes with relatively thick eyelashes, and wasn't that good enough? No. I want them pointy and huge. Couldn't you just do this with inpainting? You can inpaint with this to make them even thicker and more angular. Why would anyone use this?Well, if they want thick and angular eyelashes at any cost, obviously. As much as this model is basically a one trick pony, I'm not forcing you (thank god) to use it - so if you want thick and angular eyelashes, I'm your guy, if not, feel free to use an, admittedly, better model made by someone vastly more competent. If anyone has any ideas on how to make the eyelashes even more thick and/or angular, please let me know.I try to upload images as a normal generation resolution + remacri for sake of ease, but better quality can be achieved through img2img upscaling.My personal suggestions for settings are as follows:30 stepsEuler A/simpleCFG 61216x832/1152x896/1024x1024 (basic and boring, I know) V2.0 currently has stronger eyelashes than V3.0Neo, but V3.0Neo is a. More usable b. an easier base to build off of 3.1XXL I promise you will have stronger eyelashes. Happy eyelashing!

AegisMix

I have been in the kitchen forever, trying to maximize the ability to have huge eyelashes - for a 3, I realize I will simply have to throw together a huge dataset. Pain. The reliance on kuukansei in the 1.2 was a bit of a crutch, so I had to go back to chef-ing. The recipe for this version was <a rel="ugc" href="https://civitai.com/models/1118375?modelVersionId=1434819">Hinano </a>(x1<a rel="ugc" href="https://civitai.com/models/1057098/prsmrti-furry-style-lora">prsmrti (Sweet christ, NSFW warning)</a>, x1<a rel="ugc" href="https://civitai.com/models/1127632/artist-style-birdblitz-noobaixl-lora">blitz</a>) x 0.4 <a rel="ugc" href="https://civitai.com/models/1248959?modelVersionId=1407935">ntdmix</a>(x1<a rel="ugc" href="https://civitai.com/models/1057098/prsmrti-furry-style-lora">prsmrti</a>, x0.5<a rel="ugc" href="https://civitai.com/models/1127632/artist-style-birdblitz-noobaixl-lora">blitz</a>). I am sure more can be done, and I have a version that does more, but it specifically only works for my alternate lora I haven't released yet otherwise it just dumps furry smut. The realization that I have to make the dataset myself has been horrifying. I beg you, put 'furry' in the negative prompt if you suffer any casualties. God help us all. May the lord forgive us for our desire for a simple long eyelash.

Files with SHA256: 116c836d00aa5352c417ce98e1a987dd92e57aeed92b0c15607b9035e7dcdb92