An attempt to train a LoRA model from SD1.5. The inspiration was simply the lack of any Emiru model of any sort here.
Another attempt. used deepbooru for img2txt instead of blip. the dataset is the same. feels like the results are a bit better, but still not "ideal". will continue tinkering. used protogenX53Photorealism for the previews.
Go ahead and upload yours!
Your query returned no results – please try removing some filters or trying a different term.