

{ "subject": { "type": "young adult woman", "appearance": "fair skin with subtle freckles, green-gray eyes, defined brows, full natural lips", "hair": "short vintage auburn hair, swept back with soft volume", "expression": "calm, serious, slightly mysterious", "gaze": "looking directly into the camera" }, "wardrobe": { "top": "cream vintage blouse", "scarf": "rust-orange patterned neck scarf tied loosely" }, "setting": { "location": "inside a classic 1950s turquoise convertible", "environment": "sunny rural landscape with softly blurred hills in the background", "car": "vintage turquoise body, chrome trim, warm brown leather interior" }, "composition": { "aspect_ratio": "4:5", "framing": "tight portrait through the open car window", "camera_position": "outside the car at eye level", "foreground": "turquoise car door and chrome window frame partially framing the woman", "focus": "sharp focus on face and eyes", "background": "strongly blurred with shallow depth of field" }, "lighting": { "type": "warm natural sunlight", "quality": "soft directional sunlight with gentle facial shadows", "mood": "warm nostalgic summer afternoon" }, "visual_style": { "style": "photorealistic vintage film photography", "era": "1950s–1960s Americana", "color_grading": "warm muted tones with turquoise, amber and orange dominance", "texture": "subtle analog film grain, slight vintage softness", "lens": "cinematic portrait lens with shallow depth of field" }, "negative_prompt": [ "modern car", "modern clothing", "studio lighting", "overly polished digital skin", "plastic skin", "heavy makeup", "wide-angle distortion", "extra people", "text", "watermark", "deformed face", "extra fingers" ] }
This image was generated with GPT Image 2.5 by OpenAI. The prompt on this page is the exact text used to create it.
The example was generated with GPT Image 2.5 using: 941:1672 · Cinematic · Vintage · Portrait. Reusing these settings gives the closest match.
Copy the prompt as-is first, then replace only the subject while keeping the style, lighting, and composition keywords. Generate a few variations — outputs differ between runs.
Yes. Prompts work best as starting points — adjust descriptors, aspect ratio, or style tags to make the output your own.