stable diffusion,YUKINOlora真不错!

找了好久的站也是终于找到这个lora模型。

我自己调配好的参数可以看一下
   best quality, masterpiece, silver hair, sweater, pantyhose, blue eyes, mature,White vest sweater， office lady, (vibrant colors), solo, (dalcefo, sakimistyle:1.1), Pilyeon, (stare, realistic, lips, eyelashes:0.75), office, modern office, high-rise, colorful room, sitting on table, <lora:yukino:0.8:0,0,0,0,0,0,0.5,0.5,1.2,1.2,1.2,0.5,0,0,0,0,0> Negative prompt: (worst quality, low quality, extra digits:1.2), (mutated hands and fingers, missing fingers:1.3), dark skin, monochrome, Steps: 20, Sampler: Euler a, CFG scale: 8.5, Seed: 1036584191, Size: 600x900, Model hash: cbfba64e66, Model: 动漫模型 _ Counterfeit-V2.5_v30, Clip skip: 2, ENSD: 31337

不错，御姐大爱

请问倒数第二张是啥大模型

这个确实还可以，其实大部分lora，因为训练集的质量太低，导致崩塌率太高