WARNING: This is just a very rough test to get an estimate of how much I need to change for flux training vs sdxl/pdxl! This lora is able to reproduce the character, but not the style of the anime, so I am mainly just posting it for science reasons, not as a viable asset to make good images. Feel free to experiment with it, you might be able to get some nice results in combination with a good anime style (by prompt or a separate lora), but do not expect too much from it!
Description
EXPERIMENTAL
just trying to figure out settings for flux lora trainer, might make an actually useable version later if I get the buzz for it and time to re-caption and upgrade my dataset, but no guarantees
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.