
这是一个以PonyXLv6为底模训练的游戏画风Lora
This is a PonyXLv6 as the base model training game painting style Lora
如果你喜欢这个Lora麻烦你点个小赞或者给个Buzz谢谢您!
If you like this Lora, please click "like" or "Buzz", thank you!
注意事项:一定要写画风提示词!wrelartstyle (不写画风提示词没什么效果!)
Note: Be sure to write drawing style prompt words! wrelartstyle (No writing style prompt words have no effect!)
V2版本已经处理好了一些细节问题,但是还是存在一些问题,但后续的版本还需要看情况来定,或者后续会出Flux的版本Lora
The V2 version has handled some details, but there are still some problems, but the subsequent version needs to be determined, or the Flux version Lora will be released later
推荐参数:
Lora权重:0.5~1 或者更高(我使用一般为1,你可以设置成一个你觉得合适的参数即可)
CFG:7(你可以根据你的实际情况来定)
采样器:Euler a
采样器类型: Exponential
步数: 20~30 (v2版本的这个Lora我用20步就行了已经很好了!)
出图的宽高比随意你自己选择一个合适的就行了
Recommended parameters:
Lora weight: 0.5~1 or higher (i use it lora with 1 weight, you can set the parameter to your satisfaction)
CFG Scale: 7 (you can set it to any value you like,you like it is ok )
Sampler:Euler a
Schedule type: Exponential
Steps: 20~30 (v2 version of this Lora I use 20 steps is good!)
image width and height follow wit you common
描述:
训练词语: wrelartstyle
名称: wrelartstyle.safetensors
大小 (KB): 445817
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success