漫画 / 故事板 - Crimson Xianxia Martial Training

仙侠 · 武术训练 · 女性角色 · 古风庭院
分类:故事板与角色
更新于
提示词
Create a vertical cinematic xianxia fantasy scene in an ancient Chinese sect courtyard: a beautiful young martial heroine, {argument name="female character name" default="senior sister"}, dominates the foreground in a dynamic close-up three-quarter pose, throwing a powerful straight punch toward the viewer with her right fist foreshortened. She has vivid {argument name="hair color" default="crimson red"} hair tied in a high ponytail with an ornate gold crown hairpiece, loose strands and ribbons whipping dramatically in the wind, pale porcelain skin, sharp elegant features, intense red-brown eyes, subtle red forehead mark, and a focused fierce expression. Her outfit is an elaborate dark black-and-red armored cultivator costume: sculpted leather-and-metal chest armor, gold filigree trim, layered bracers, chain ornaments, dangling red beads, translucent crimson silk sashes, and a short flowing battle skirt with sheer fabric layers. Red energy embers and smoky magical trails spiral around her punch and across the lower frame. In the background, exactly one male observer, {argument name="male character name" default="Zhao Wuyou"}, stands slightly out of focus on the stone steps, watching her practice; he is a handsome young cultivator with long black hair in a high half-up style, wearing a dark charcoal-black embroidered robe with metallic patterns and layered sleeves, calm and serious. The setting is a misty grand immortal palace courtyard with stone stairs, carved railings, ancient temple architecture, curved roofs, distant pavilions, soft blue-gray morning haze, and shafts of cool daylight. Use an ultra-detailed high-end fantasy illustration style with realistic skin, intricate costume textures, shallow depth of field, dramatic wind motion, glowing red particles, cinematic lighting, strong foreground-background depth, and a vertical 9:16 composition. No text, no watermark, no modern objects, no extra characters.使用提示词GPT 内容评估
AI 自动评价,仅供参考
模型:GPT-6-luna 综合:8.7 / 10 提示词质量:8.4 / 10 图片表现:9.1 / 10 图文一致:8.7 / 10 画面冲击力强、与提示词高度吻合,适合作为上线主视觉;若需体现“漫画 / 故事板”,建议进一步明确分镜形式。
评价于