{ 'meta': { 'system_instruction': 'Create a high-end fashion magazine cover. Use the uploaded face for strict identity preservation.', 'aspect_ratio': '3:4', 'quality': 'ultra-deta...

Use the uploaded photo as the base. Keep the person’s **face, eyes, hairstyle, makeup, outfit, and overall identity** clearly recognizable and photorealistic. Do NOT turn the whol...

{ 'scene': { 'description': 'Stylish restroom dressing area captured via mirror.', 'environment': 'Warm wood-laminate stall doors and soft countertop reflections.', 'mood': 'Shy-pl...

A Subject: me (use uploaded face — 100% facial accuracy, no changes). with no extra brightness on the face, keeping my natural dark facial tone intact. dreamlike, movie-quality au...

参考图1的面部特征，生成全身工作室肖像：一位英俊的年轻东亚女性坐在浅紫色背景前的地板上，穿着舒适的超大号薰衣草色粗针织毛衣、白色裙子和白色袜子，深情地抱着一个大型三丽鸥库洛米毛绒玩...

{ 'image_prompt': { 'subject': { 'face_preservation': true, 'description': 'A beautiful young woman kneeling inside a cartoon-style monochrome brown room. Her facial features must ...

{ 'pipeline_configuration': { 'job_type': 'img2img_transformation', 'meta_tags': ['macro', 'beauty', 'soft_focus', 'realism'], 'input_reference_handling': { 'preservation_rules': {...

Present a clear, 45° top-down isometric miniature 3D cartoon scene of [CITY], featuring its most iconic landmarks and architectural elements. Use soft, refined textures with reali...

A whimsical, highly detailed photograph of a tiny artist (miniature person, about 5cm tall) standing on a wooden desk, painting on a large canvas. The canvas is enormous compared t...

基于我提供的【参考人物照片】生成一张“穿搭拆解信息图”（Outfit Breakdown）。 【硬性要求】 人物居中、清晰、与参考照片一致（同一人同一穿搭）。 四周用标注卡片 + 引线箭头拆解单品。 每...