search

hollowstrawberry / kohya-colab

Accessible Google Colab notebooks for Stable Diffusion Lora training, based on the work of kohya-ss and Linaqruf
GNU General Public License v3.0
564 stars 79 forks source link

Please, can you increase maximum dim and alpha for 128. And add dreamsharper Xl as a default model option #158

Open pedroxceeed opened 1 month ago

pedroxceeed commented 1 month ago

Please, can you increase maximum dim and alpha for 128 (the current limit is only 32)

And add dreamsharper Xl as a default model option ( I know there is a option for dowload another models but drive crash ang get error when I add a civitai link

gwhitez commented 1 month ago

Please, can you increase maximum dim and alpha for 128 (the current limit is only 32)

And add dreamsharper Xl as a default model option ( I know there is a option for dowload another models but drive crash ang get error when I add a civitai link

use this: https://huggingface.co/stablediffusionapi/dreamshaper-xl

hollowstrawberry commented 1 month ago

I can't approve of 800 MB Loras, there's no reason to train with the dim that high.

I'll probably add dreamshaper as a default option though.