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 + YandereNeoXL 200k (I like the heavy one, what can I say) for sake of ease, but better quality can be achieved through img2img upscaling.
My personal suggestions for settings are as follows:
30-40 steps
Euler A/SGM uniform (I just think its neat)
CFG 6
1216x832/1152x896/1024x1024 (basic and boring, I know)
There are probably better options even, but I never claimed to be good at any of this. Suggest and I will improve!
Happy eyelashing!
*Forgot to mention that the side-goal of this project is more anime-style coloring as opposed to the semi-real and 2.5d aesthetics, and I forgot to mention that because it's about the lashes, ???? it!
Description
This version doubles down on the original intention of the model, making the eyelashes and outlines even thicker. I think we've got some really good eyelash results this time, but the effort will continue after I finish the new version of my other LoRa (oooo spoilers in preview images!). I'd tell you what went into this, but, because I'm comically stupid, I fried the metadata! So from the top of my head, let's try to remember what I put in here. Of course, the main new addition is the base model + 0.4 of the wonderful model KuukanSei. That model is a real work of art & other than a certain comment, inspired me to iterate again before doing any more LoRa work. Upon that, after some model testing, I decided to add in a 0.3 of the incredibly pretty Hinanomix to spruce up the pallet a bit. On top of that, I drummed my fingers for a bit - it was almost there, but not quite. So, I spliced in this LoRa because I thought it was look interesting, and there we had it - the completed work. There was a 1.2b, but I canned it entirely and decided to call this good enough for now. Suggestions would be appreciated as ever, onwards and upwards gentlemen!
Most of the images are the exact same prompt/seed as the v1
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.