This is a test to see if reducing the dimension to 8 is enough to use non-ideal images. If this works, then i might not need to use high qualty anime style images to get an anime output, just drawings in general.
Trained on 24 epochs, 612 steps and 3 repeats.
Description
A version of the very rare front to back configuration of two torsos conjoinment. There is not enough pics to use, so i had to improvise with edits.
This lora needs (conjoined:1.4), (multiple girls:1.2), and a 1.3 weight.
This lora mostly works, and it works better than i expected given the bad edited training images i used, but it feels limited. And it seems it needs img2img and some manual corrections after generating. Use it if you feel that you can turn it into something good, but don't expect much.
FAQ
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.