兽耳酱kemomimi57_ponyB_LORA
所使用的底层大模型是:ponyDiffusionV6
LORA推荐权重:0.8左右
触发词: 1girl, kemomimi-chan \(naga u\), solo, tail, animal ears, skirt, sleeves past fingers, sleeves past wrists, fox ears, thighhighs, fox tail, shirt, pleated skirt, green shirt, neck bell, hair ornament, bell, jingle bell, hair bun, purple skirt, fox girl, white background, ribbon-trimmed legwear, white thighhighs, bangs, collar, single hair bun, long sleeves, brown collar, flat chest, blonde hair, hair between eyes, orange necktie, ribbon trim, sailor collar, necktie, animal ear fluff, white sailor collar, shadow, full body, red eyes, closed mouth, standing, brown footwear, blush, sidelocks, looking at viewer, long hair
生成参数推荐:
采样方式:DPM++ 2M Karras
迭代步数:20
高分辨率修复开
重绘采样步数:20到30之间
重绘幅度:0.5
放大算法:R-ESRGAN 4x+ Anime6B
放大倍率:1.5
宽度:768
高度:960
ADetailer启用 (要开)
ADetailer model: mediapipe_face_full
负面提示词: (worst quality, low quality:1.4),[:(mutated hands and fingers,one hand with more than 5 fingers, one hand with less than 5 fingers):0.8],(Multiple people),(watermark,username,artist name:1.2),lowres,bad anatomy,bad hands,text,error,missing fingers,extra digit,fewer digits,cropped,worst quality,low quality,normal quality,jpeg artifacts,signature,blurry
Animal ear sauce k bad mood mimi57 _ ponyB_LORA
The underlying large model used is: ponyDiffusionV6
Lora recommendation weight: about 0.8
Trigger word: 1girl, k is in a bad mood, mimi-chan\ (naga u\), solo, tail, animal ears, skirt, sleeves past fingers, sleeves past wrists, fox ears, thighhighs, fox tail, shirt, pleated skirt, green shirt, neck bell, hair ornament, bell, jingle bell, hair bun, purple skirt, fox girl, white background, ribbon-trimmed legwear, white thighhighs, bangs, collar, single hair bun, long sleeves, brown collar, flat chest, blonde hair, hair between eyes, orange necktie, ribbon trim, sailor collar, necktie, animal ear fluff, white sailor collar, shadow, full body, red eyes, closed mouth, standing, brown footwear, blush, sidelocks, looking at viewer, long hair)
Generate Parameter Recommendations:
Sampling mode: DPM + + 2M Karras
Iteration steps: 20
High Resolution Repair Development
Redraw sample steps: Between 20 and 30
Redraw amplitude: 0.5
Amplification algorithm: R-ESRGAN 4x + Anime6B
Magnification: 1.5
Width: 768
Height: 960
ADetailer Enabled (To be On)
ADetailer model: mediapipe_face_full
Negative cue words: (Worst quality, low quality: 1.4), [: (mutated hands and fingers,one hand with more than 5 fingers, one hand with less than 5 fingers):0.8],, multiple people) (watermark,username,artist name:1.2),lowres,bad anatomy,bad hands,text,error,missing fingers,extra digit,fewer digits,cropped,worst quality,low quality,normal quality,jpeg artifacts,signature,blurry)