Garland Style V2.3 is out!
This model is trained base on my own artworks, used DreamShaperV8. You can find my art at pixiv:
https://www.pixiv.net/users/18736174
V2.3版本是基于我的个人绘画作品训练的，基础模型是DreamShaperV8，并融合了一些其它模型。你可以在PIXIV查看我的个人页面：
Description
The first version.
FAQ
Comments (2)
it's perverse and creative. very interesting model you've made here, i wonder what process you used, as well as the possibility of adding more range in future models
I use dreambooth to train an original style model, that depends on trainning parameters and source images. Then to change the in/out layer weight of alpha/beta with supermerger extension(other tools is the same), to balance the differences of understanding/generate ability of models. Not too difficult but my own aesthetic disposition leads.😄 About the future model, I'm waiting for the SD2.2, the preview of it seems to be more compatible with hands, I guess feet though, that attracts me a lot.