
This is an Artstyle of several artist I found on several website they were similiar styles but rarely any photos so I build the source materials after the character Ive seen online.
recomended weight: 0.8-0.6
Use any Anime Checkpoints like Aingdiffusion or Mistoon etc.
I dont recommend the v1 at all it has several problems cause its trained more on character parameters instead of that a Style should be. Thats why it often will try to reproduce the chars in the source material onto the loras u use with it or chars you craft.
This version is trained on better parameters that a Style actually should be.
Since its still abit away from what I want to achieve I may consider trying myself on a sd checkpoint
Loras used in the pictures:
Heroinex by NeuralDavinci: https://civitai.com/models/198436/mysterious-heroine-xx-or-artoria-fate-series-or-neural-da-vinci
Sheba by me: https://civitai.com/models/261384/sheba-fate-grand-order
描述:
I recommend a weight of 06-1
训练词语:
名称: Korean_AI_bot-000009.safetensors
大小 (KB): 36977
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success