Manmaru mix
I don't speak English so I'm translating at DeepL. Let me know if the English is weird.
This model was adjusted so that the hand-drawn look would fade and the background
would come out clearly, while keeping the characterization of the Thumbelina Model.
このモデルはThumbelina Modelのキャラデをそのままに、
手描き感を薄くし、背景がくっきり出るように調整しました。
Notice / 告知
Onigiri Mix is now available, with the same lovely design as Manmaru,
It is more suitable for depicting hands and backgrounds. Please give it a try.
Onigiri Mixをリリースしました。Manmaruと同じように可愛らしいデザインで、
より手や背景の描写に強いです。お試しください。
Tips / 参考
This model works well with LoRA, which enlarges the eyes. (e.g. bigeye)
目を大きくするLoRAと相性がいいです。(例:bigeye)
Update History / 更新履歴
V2
The background is clearer than in V1. Also, the hands should be just a little bit better...
After using it myself for a while, I noticed that the face of the output is cuter than V1. That's my subjective opinion.
背景がV1よりよりくっきりしました。あと、ほんのちょっとだけ、手の描写がましになっているはず・・・
しばらく自分で使ってみて、顔がV1よりかわいくなっていることに気づきました。私の主観ですが。
Merging Method / マージ方法
V1
A:Thumbelina_v20.safetensors
B:nostalgiaMix_nostalgia.safetensors
Sum Merge : -0.6 (THIS IS "minus"!! マイナスです!!)
->tmp.safetensorsA:flat2DAnimerge_v20.safetensors
B:tmp.safetensors
C:featurelessMix_v10.safetensors
MBW:
alpha:0.25,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5,0.5
beta:0.5,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
V2
A:flat2DAnimerge_v20.safetensors
B:cocotifamix_v20This.safetensors
Sum Merge : 0.5
->tmp.safetensorsA:manmaruMix_v10.safetensors
B:tmp.safetensors
MBW:
alpha:0,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,1,1,1
Origin of the model name / モデル名の由来
"manmaru" = "まん丸", means "perfectly round".
Because the eye shape output by this model is round.
このモデルで出力される目の形がまん丸なので、"manmaru"。