
The model is used to optimize the details of character clothing, mainly for bikinis, cheongsam, evening gowns, and their mixed clothing. "But not limited to this function, the original version 1.0 of Lora can generate clothing, while the main functions of the later version 2.0 are to optimize clothing details, overall screen feel, and correct body misalignment.".Can be used for mechanical elements.
The weight value should not exceed 0.5 after version 2.2.
Version 3.0 should not exceed 0.4
Fixed 2.5d character weight usage of 0.05~0.1
该模型用于优化人物服装细节,主要用于比基尼,旗袍,晚礼服,以及他们的混合服装,但不限于该功能,该lora原始1.0版本可以生成服装,2.0后版本主要功能在优化服装细节,整体画面感,修正身体错位,可用于机械元素,
该lora貌似可以自主绘画,某种情况下会抛弃大模型绘制自己喜欢的内容,遇到这种情况请打他的头,
2.2版本以后 权重值不宜超过0.5,
3.0版本不宜超过0.4.
修正2.5d人物权重使用0.05~0.1
描述:
Adding a lock enhances the adaptability of the model's actions, but reduces stability
添加了locon增强了模型动作适应性,但降低了稳定性
训练词语: mix_lifu\(bikini\),mix_lifu\(ceremonial_dress\),mix_lifu\(cheongsam\),mix_lifu\(mix_cheongsam_bikini\),mix_lifu\(mix_ceremonial_cheongsam\),mix_lifu\(mix_ceremonial_cheongsam_bikini\)
名称: mix_lifu3.1.safetensors
大小 (KB): 340793
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success