V1 is trained on 60 images and captioned with Civit Ai's auto caption feature. Settings left at default.
V2 is trained on more images but without captions. May be better or worse.
V3 turned out really good so far. Would love to see what you make with it.
Prompting: "Highly detailed traditional artwork by Ayami Kojima of..."
Feedback is welcome.
描述:
Trained on more images with more epochs, let me know if seems better or worse.
训练词语:
名称: Ayami_Kojima_V2.safetensors
大小 (KB): 18806
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
