thumbnail: >-
https://media.discordapp.net/attachments/1002437703192821910/1073391952977989632/thumb.png
license: creativeml-openrail-m
language:
- en
library_name: diffusers
pipeline_tag: text-to-image
VOXO
Merged model by junjuice0.
This model was originally created just for me, so I am not after quality and please don't expect too much.
I may release finetune version of this model in the future, but only God knows if I am willing to do it until then.
VOXO-Vtuber (VOXO-v0-vtuber.safetensors)
This model can generate vtubers for Hololive and Nijisanji. Some vtubers may or may not come out well. It is recommended to give the name a weight of about 1.2 (e.g. (ange katrina:1.2))
RECOMMENDED
It is recommended to use TIs such as bad-images or bad-prompt for negative prompts. Also, quality prompts (e.g. masterpiece, high quality) are not required. The use of highres. fix may change the painting considerably, use according to your preference.
HOW TO USE
The usage is the same as other diffusion models, and it would be easier to read other people's explanations than mine here.