Type | |
Stats | 445 160 |
Reviews | (37) |
Published | Apr 9, 2024 |
Base Model | |
Training | Steps: 5,000 Epochs: 10 |
Usage Tips | Clip Skip: 2 Strength: 0.8 |
Hash | AutoV2 4B1E0358A7 |
This is a cartoon character/3D anthropomorphic animal LoRA with good generalization. It is trained from a total of 159 various characters and animal (anthropomorphic) materials. It is especially suitable for IP design. The training base model ReVAnimated is also recommended to use this base model for drawing. Of course, you can choose other ones.
Parameter reference example picture (there is no global trigger word, of course there is a built-in peeping Easter egg mode, the trigger word: peeking out, looking at viewer).
Negative prompts: badhandv4,EasyNegative,worst quality,low quality,normal quality,lowres,bad anatomy,bad hands,watermark
这是一款泛化很好的卡通人物/3D拟人动物LoRA,总共由159张各种的人物、动物(拟人化)素材训练而成,特别适用于自媒体。训练底模ReVAnimated,也推荐使用这个底模生图,当然选择其他的也没事。
参数参考例图(没有全局触发词,当然内置了个偷窥彩蛋模式,触发词:peeking out, looking at viewer)。
负面提示词:badhandv4,EasyNegative,worst quality,low quality,normal quality,lowres,bad anatomy,bad hands,watermark