返回 Prompt 库
“Caught Between Soft Smiles Sleepy Eyes and
“Caught between soft smiles, sleepy eyes, and little moments that feel like home ☁️🫶✨” GPT Image 2 on ChatGPT Subject & Composition Format: A 4-panel grid layout (2x2 collage) sho

Prompt
“Caught between soft smiles, sleepy eyes, and little moments that feel like home ☁️🫶✨”
GPT Image 2 on ChatGPT
Subject & Composition
Format: A 4-panel grid layout (2x2 collage) showing four different candid expressions of the woman from image.png.
Setting: A cozy, intimate "blanket fort" environment with soft, white linen sheets draped overhead to create a diffused, tent-like glow.
Expressions: One panel showing a shy smile with a hand on the cheek, one laughing with a hand over the mouth, one playful wink with tongue out, and one soft pouty expression.
Attire: A simple pastel tie-dye camisole with thin spaghetti straps, one strap casually slipping off the shoulder.
Technical Specs & Aesthetic
Lighting: Soft, natural morning light filtered through white fabric, creating gentle shadows and high-key luminosity.
Camera: Shot on a high-end mirrorless camera using a 35mm wide-angle lens to capture the close-up, intimate "selfie" perspective.
Settings: Large aperture of f/1.8 for a shallow depth of field, keeping the facial features sharp while the white bedding in the background softly blurs.
Style: Pinterest-inspired aesthetic with a hint of 90s retro film grain. The color palette should be dominated by airy whites and soft pastel tones.
Materials: Realistic skin textures with subtle natural highlights and soft fabric PBR (Physically Based Rendering) for the linen sheets.
Signature Detail
Include the signature "XSydneyFan" in a small, elegant cursive font in the bottom right corner of the final grid.
2:3ar
3D RenderPhotoTypographyCinematicVintage PrintHero PortraitNature LandscapeSoft LightTop DownSocial Media Post
如何使用这个 Prompt
- 用上方的复制按钮复制完整提示词。
- 把方括号 [ ] 里的占位内容替换成你自己的主体、场景或细节。
- 粘贴到 Gemini / Nano Banana Pro 生成,或直接点上面的「试试这个 Prompt」在本站免费生成,再对照示例图微调用词。
风格说明
- 3D Render
- CGI 质感:表面平滑、全局光照可控,纵深像 Blender/C4D 渲染而非照片。
- Photo
- 要求模型输出照片级真实感:自然光线、可信的皮肤与材质、类似相机的景深,而不是插画或 3D 质感。
- Typography
- 文字本身是画面的一部分。Nano Banana Pro 这类新模型能渲染清晰的美术字,但提示词里要把具体文字写短、并用引号标出。
- Cinematic
- 电影感渲染:宽画幅构图、强烈的光比、调色(常见青橙对比),画面自带叙事氛围。
- Vintage Print
- 旧印刷品质感:颗粒、网点、褪色油墨、复古字体,唤起上世纪海报和老杂志的味道。
- Hero Portrait
- 把单一主体拍成主视觉:紧凑构图、有力的眼神或姿态,背景简化,让人物撑起整张画面。
- Nature Landscape
- 广阔的自然风光构图:层次纵深、天气与时间(晨昏)描述承担大部分氛围表达。
- Soft Light
- 柔和低对比的光线:阴影轻柔、肤色讨喜,类似窗光或阴天的效果。
- Social Media Post
- 为社交信息流(Instagram、X、小红书)优化的构图与画幅:主体突出、预留文案位置、配色在小屏幕上依然醒目。
常见问题
这个 Prompt 怎么在 Gemini(Nano Banana Pro)里使用?
点击上方的复制按钮拿到完整提示词,粘贴到 Gemini 或任何支持 Nano Banana Pro 的工具里,再对照本页示例图检查效果。也可以直接点「试试这个 Prompt」在本站生成——每天有免费次数,无需注册。
可以修改「“Caught Between Soft Smiles Sleepy Eyes and」这个提示词吗?
可以。方括号 [ ] 里的内容是占位参数,本来就应该替换成你自己的主体或细节;主体、光线、风格关键词也都可以自由调整,措辞上的小改动往往带来明显不同的结果。
这个提示词适用于哪些 AI 模型?
它最初是为 Google 的 Gemini 图像模型(含 Nano Banana Pro)分享的。大多数文生图模型也能理解,但构图和文字渲染质量会因模型而异。





