The model used more than 750 high-quality original pictures for training.
The Version 3.0 has more character facial shapes and body movements, and the details are even better.
Recommendation Weight：0.6~1.0,
Recommendation VAE：animevae.pt
Recommendation Model：scmix_NSM.ckpt
--- --- --- --- --- --- --- ---
该模型使用了超过750张高质量的原始图片进行训练,
3.0版本LORA拥有更多的人物脸部造型和肢体动作，细节表现更加优秀,
推荐权重：0.6~1.0
推荐vae：animevae.pt
推荐模型：scmix_NSM.ckpt
描述:
Increase FP32，Update to v1.1，The difference between v1.0 version is not very big, perhaps some details are slightly better. :p
This is my first LoRA : Satoshi_Urushibara_v11_fp32
うるし原智志 (漆原智志)
This Lora use 200+ picture
I'd recommend using a weight value 0.6~1.0
Orangemix.vae is good for this model
For my tests I have used scmix_NSM
训练词语: Satoshi Urushibara,,1girl
名称: Satoshi_Urushibara_v11_fp32.safetensors
大小 (KB): 295038
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
