Trained on a smaller dataset of 150 fully rendered works by Raiou. https://twitter.com/raiounsfw
I recommend "realistic" as the negative prompt.
Unlike the SD1.5 version, it's mostly trained on his newer stuff, so expect clean linework. I might train a sketchy/soft lines model later, idk, let me know if that's something you want.
Description
Trained on a smaller dataset of 150 fully rendered works by Raiou. Unlike the SD1.5 version, it's mostly trained on his newer stuff, so expect clean linework. I might train a sketchy/soft lineart model later, idk, let me know if that's something you want.
FAQ
Comments (2)
Good lora, can do with cleaned watermarks.
Yeah, I didn't realize while pruning the dataset that most of the ones I kept had the watermark. Whoops!