【AI绘画】测试记录

模型：dalcefo_v3_painting

https://civitai.com/models/5396/dalcefov3painting

乃琳人物lora：https://civitai.com/models/10530/asouleileenv20

跑出来是这样

接着降低了下lora权重

要增强人物特征，又使用了embeddings模型（.pt结尾那个）

效果显著！！

然后微调、修改、完善tag，本想着顺便再放大一下

最后效果也还行，随便贴几张稍微好点的

Prompt:((realistic)),(1girl),(nailin),painting,(white hair),bare shoulders, chromatic aberration,((fox ears)), ((cinematic light)), (finely detailed face), (portrait), (long, flowing)

, cinematic light, disheveled hair, pastel colours, muted color,(enchanting:1),bestquality,(hair floating),(extremely beautiful detailed blue eyes),((red eyeshadow)),(closeup),long eyelashes,closed mouth,dynamic angle,

,intricate detail, fairy,red eyeliner,((red makeup)),masterpiece,collarbone,an extremely delicate and beautiful girl,(Winter snow),(long hair flowing with the wind),snow,from side,night,looking at camera

Negative prompt: EasyNegative, monochrome, (worst quality, low quality:1.4),lowres,badanatomy,badhands,badfoots,wrong,badfingers,text,error,missingfingers,extradigit,fewerdigits,cropped,worstquality,lowquality,normalquality,jpegartifacts,signature,watermark,usemame,blury,badfeet,futa,futanari,small_penis,yaoi,huge_breasts,large_breasts,three legs,wrong hand,wrong feet,wrong fingers,deformed leg,abnormal,malformation

Size: 768x512

Steps: 随意

Sampler: Euler a, DPM++ 2M Karras，DDIM，DPM++ 2M Karras

CFG scale: 7, 

LORA模型权重 0.1

本来正面还有（white fox）来着，跑了这么久要么不出现要么两者融合，还是不折磨我这4G小显卡了（懒得搞了）