AI art "あくたん:制服" featuring #湊阿庫婭 and #Hololive Production
生成新作品

あくたん:制服

#湊阿庫婭#Hololive Production
更多精彩作品

留言

minato_aqua, minato_aqua_(school_uniform), A VTuber with long, multicolored hair—blue and pink streaked—wears a neatly fitted Japanese school uniform, consisting of a white blouse with a loosened ribbon and a pleated skirt. She stands in a relaxed yet dynamic pose, slightly turning her body toward the viewer with a gentle smile and soft gaze. The watercolor-style illustration features soft lighting and subtle paint splatter effects, enhancing the refined anime aesthetic. The background is a simple, abstract setting with shallow depth of field, keeping focus on her delicate expression and the intricate details of her uniform.
顯示參數
原始提示詞
looking at viewer, (watercolor:1.5), full body, 1girls, minato aqua, (vtuber), soft lighting, soft focus, clean illustration, refined anime style, school uniform, (Japanese school uniform OR sailor uniform OR blazer uniform OR academy uniform OR modern school uniform), (white blouse OR sailor top OR buttoned blazer OR cardigan layered uniform), (pleated skirt OR straight skirt OR short skirt OR long skirt), (neatly worn OR slightly loose uniform OR casual uniform style OR ribbon loosened OR open collar OR cardigan tied around waist), (dynamic pose OR relaxed standing OR walking pose OR slight turn pose), natural expression, gentle smile OR calm expression OR soft gaze, simple background OR abstract background OR school corridor OR classroom OR street background, subtle paint splatter, shallow depth of field, high quality, official anime style, score_9, score_8_up, score_7_up
品質標籤
high quality, year_2025, year_2024, ﹝提示詞﹞
尺寸
720 x 1120
取樣步數
25
取樣方式
Euler
CFG Scale
4.5
負面
bad quality, low quality, worst quality, lowres, nsfw, jpeg artifacts, scan artifacts, mosaic, cropped, error, fewer digits, bad reflection, bad composition, bad anatomy, bad hands, bad fingers, missing fingers, extra hands, extra legs, excess fingers, light particles, artist name, text, watermark, username, copyright name

使用的模型和 LoRA