Lib Exclusive | Simple Water-based E-commerce Background-SDXL
v1.0






Bottom model: basic algorithm _ XL (recommended large model: Wanxiang _ realistic (independent), I feel that the effect is very good)
Wanxiang_Multi-class model (independent):
https://www.liblib.ai/modelinfo/4b17e39e13a449648cf9bc306ca3ee39Recommended LoRa weight: 0.6-0.8 Trigger words
Recommended hint words:
a round flat podium, water, sky, cloud, depth of field, flower, moonlight , night, flower petals floating on the water surface, mountains in the distance, smooth floor, use your imagination to adjust LoRa weights,
use otherhints Fill in
negative cues: NSFW
, nude, naked, porn, ( worst quality, lowquali-ty: 1.4), distorted, discouraged: 1.3), croppedout of frame, slightly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbscloned face, (mu-tated hands and hands and fingers: 1.4), disconnected fingers, extra legs, fused fingers, toomany fingers, long neck, mutation, mutated, uglydisgusting, amputa-tion, blurry, jpeg artifacts, watermark, watermarked, text, signature
, Sketch sampling method: DPM++SED Karras iteration steps:
Recommended size: 768×1024/800×1200
(3:4/2:3)Cue Guidance Factor (CFG)
): 7.0 Kneel and ask the big guys to return the picture!
