Bitchko - jk bitchi no shiboretai - CivArchive (CivitAI Archive)

Bitchko - jk bitchi no shiboretai - version_AL

NSFW

This lora objective is to help in the generation of images of Bitchko, from the "jk bitchi no shiboretai" serie.

There aren't that much images of her around, so the results are a bit limited, but overall the lora i've trained seems to work decently enough.

Let's talk about those more in detail, there are currently 3 version of the lora in beta testing, and this page will be updated once the test are finished, for more detail about the versions you can check the proper section of the page.

p.s. We spent a lot of buzz trying to produce a decent lora, so consider supporting us by posting your images, or donating buzz, so we can keep training models.

When there will be a good amount of better images i will train a definitive version, till then, i hope you will enjoy my work.

Note: each version has a different use/purpose, there's no best one.

Description

This version was created using AnyLora ckpoint.

It has some advantages and some issues compared to the version_CA.

This version is more flexible, it allows you to vary the outfit more easily, and overall has no major issues with poses (even toh the fallatio gesture issue still persist).

This version has 2 keyword:

Bitchko  <-- for the character
from_jk_bitchi_no_shiboretai  <-- this affect the style (but somewhat help)

The strenght of the lora should be between 0.7 - 1; personally i suggest the use of it at 0.8; but there are exceptions.

This version can affect the following prompt:

school uniform, gray_skirt, scrunchie <-- for the outfit
kogal, gyaru <-- for the concept of the GAL
makeup <-- this enhance the mascara around eyes, but it can ruin the output

P.s. The scrunchie tag might, change occasionally the hair of the character adding 1 ponytail, so you might wanna put ponytail in the negative tags (this occour pretty rarely, but it must be mentioned)

Those are some negative tags i'd suggest:

text, kanji, speech bubble <-- those can fix some images with text
eating <-- it might fix the awkard mouth in some images
open clothes <-- it's suggested when trying to generate sfw images

This version is overall more flexible, but it has a problem when dealing with 768*768 images. It will try to generate faceshots of the character (not too similar between them, but it is still an issue), but it can be fixes by changing the aspect ratio, or via promt with some tweak.