返回提示词库
建筑空间 / PROMPT

90 年代公寓场景参考板

从公开案例收录的建筑空间提示词,适合参考其画面结构、风格控制和细节描述。

适用模型GPT Image 2
已实测灵感来源于 awesome-gpt-image-2 案例 381;当前提示词与效果图由 Fastgoing 使用 GPT Image 2 实测重新制作。
Fastgoing 实测结果TESTED / RESULT
Fastgoing 使用 GPT Image 2 按当前提示词生成的独立实测效果。

Fastgoing 使用 GPT Image 2 按当前提示词生成的独立实测效果。

查看原图
完整提示词COPY / USE
{
  "type": "scene reference board — 90s apartment living room, cinematic night",
  "style": "cinematic film photography, 35mm grain, warm amber shadow fill, deep chiaroscuro lighting, hyper-detailed interior, production design reference quality",
  "layout": {
    "main_panel_center_left": {
      "label": "CAMERA A — FRONT VIEW",
      "scene": "Wide shot, L-shaped tan sectional sofa, grey knit throw blanket, wooden coffee table (remote, mug, ashtray, Rolling Stone stack), lava lamp left, table lamp right, rain-streaked city window behind, Nirvana poster left wall. 35mm grain."
    },
    "main_panel_center_right": {
      "label": "CAMERA B — REVERSE VIEW",
      "scene": "Wide reverse from behind sofa. CRT TV prominent right, grey static screen. Tall bookshelf, VHS tapes. Cool blue backlight from window behind camera. Deep shadow."
    },
    "prop_strip_bottom": "6 close-up tiles: 1. LAVA LAMP — chrome base, blue-green wax blobs; 2. COFFEE TABLE — remote, mug, ashtray, magazines; 3. NIRVANA POSTER — black smiley face, wall texture; 4. CRT TELEVISION — static screen, VHS stack; 5. WINDOW/RAIN — city bokeh, water streaks; 6. THROW BLANKET — sofa corner, worn upholstery",
    "top_right_inset": "SOURCE REF thumbnail — original photo",
    "footer": "2700K PRACTICAL · 4100K CITY NIGHT · 24mm · 35MM"
  },
  "background": "deep charcoal #1a1a1a, thin white separators",
  "dimensions": "wide landscape 3:1, high resolution"
}

复制后,可参考下方“建议替换”项目改成你的主题;带【方括号】的内容请优先替换。

PROVENANCE / 来源与验证

它从哪里来,是否完成实测?

灵感来源于 awesome-gpt-image-2 案例 381;当前提示词与效果图由 Fastgoing 使用 GPT Image 2 实测重新制作。

GPT Image 2 · 2026-08-18 实测
HOW TO USE

不用学提示词工程,也能改成自己的。

建议替换

01空间类型02建筑风格03材质04时间与光线

使用小贴士

先只替换变量跑一次,观察最不符合预期的部分;第二轮只补充那一项约束,通常比一次塞进很多要求更稳定。