Illustriousベースのモデルを使ってマージしました
バージョンごとに大きく変わっているかもなので
お好みのバージョンをお使いください
(google translation)
There may be significant differences between versions, so please use the version you prefer.
branchBはv13.0の時で使用したモデルの最新版で再作成したもの
サンプルイメージの設定は以下。reforge使用
The sample image settings are as follows:
Sampler:DPM++ SDE
Schedule type:Automation(Karras)
[Normal]
Step:6~12
CFG Scale:4~6
[DMD2 ver.]
Step:4~
CFG Scale:1
Description
マージに使うモデルを減らしました。LoRAをマージしました。
Reduced the number of models used for merging.
LoRA has been merged.
FAQ
Comments (3)
Now it's really coming together!
Yes you're right, 4.0 degraded. Only regret that face looks like an ai-generated ugliness of commons, a lite pity. Of course, if face like 3.4 or improved that'll be a good one.
Any way, Thanks for this work!
oh, I forgot to say, some are improved, Adding some loras are looking better for my use, The face ai-looking problem just appeared without loras.












