hololive-diffusion, a stable diffusion 2.1 768x768 model.
This model is a fine-tune of Waifu diffusion 1.5.
hololive-diffusion is released under the Fair AI Public License 1.0-SD (
https://freedevproject.org/faipl-1.0-sd/). If any derivative of this model is made, please share your changes accordingly. Special thanks to ronsor/undeleted (
https://undeleted.ronsor.com/) for help with the license.
When creating characters owned and copyrighted by cover corp, you may not use this model or any of its outputs in a way that is counter to Hololive's fanart guidelines:
https://hololivepro.com/en/terms/The model was trained using the following names. These aren't necessarily hololive vtubers, and the characters with less popular artworks will be harder to get correct.
