wlop画风lora
XL风格模型 基于animagineXLV31_v31训练
dim:8 alpha:4
LORA推荐权重:0.8左右
触发词:realistic,blurry
采样方式:DPM++ 2M Karras
迭代步数:20到30之间
高分辨率修复开
重绘采样步数:20到30之间
重绘幅度:0.5、0.75
放大算法:R-ESRGAN 4x+
放大倍率:1.5、2
宽度:768
高度:960
ADetailer启用 (要开)
ADetailer model: face_yolov8n.pt
负面提示词:(signature, watermark, logo, text, username, artist name,web address:1.2),(low resolution),(logo),nsfw,lowres,(bad),(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],
WLOP style LoRa
XL style model is based on animagineXLV31_v31 training
dim:8 alpha:4
LoRa recommended weight: about 0.8
Trigger words: realistic, blurry
Sampling method: DPM++ 2M Karras
Number of iteration steps: between 20 and 30
High-resolution fix open
Redraw sample steps: between 20 and 30
Redraw amplitude: 0.5, 0.75
Amplification algorithm: R-ESRGAN 4x+
Magnification: 1.5, 2
Width: 768
Height: 960
ADetailer Enabled (to be on)
ADetailer model: face_yolov8n.pt
Negative Prompts: (signature, watermark, logo, text, username, artist name,web address:1.2),(low resolution),(logo),nsfw,lowres,(bad),(worst quality, low quality:1.4),[:(mutated hands and fingers,one hand with more than 5 fingers, one hand with less