



📱 Retro old phone casual shots in a rural village, 9:16; an 18–22 year old, clearly adult beautiful East Asian woman, about 1.75 m tall, delicate features, fair cool-toned skin, long black hair, tall and slender model figure, chest visually around natural E cup. Red top with white polka dots, deep V halter top × short denim shorts, natural light makeup, sweet and shy temperament. 🏡 In front of an old rural house, bamboo rocking chair, dirt ground, old brick wall with overgrown plants, holding watermelon and an iron spoon in hand. Dim yellow dusk light, dark soft-focus, slight blur, high ISO noise and coarse film grain, like a real-life snapshot accidentally captured by an old phone. Random action pool: 🍉 Sitting on bamboo chair, head down scooping watermelon, looking up with a shy smile 🌬️ One hand brushing hair aside, the other holding a watermelon, turning head against the wind 🪑 Curled up on the rocking chair, legs naturally tucked in, head down in a daze 🥄 Raising the iron spoon ready to eat watermelon, suddenly caught by the camera 🌿 Leaning sideways against the chair back, looking at the plants outside the yard 🙆🏻♀️ Sitting and stretching, long hair blown messy by the wind 🍉 Holding watermelon with both arms close to chest, tilting head slightly to look at the camera 🫧 Half-lying on the bamboo rocking chair, languid and absent-minded expression 🎲 Freely improvise old house details, wind-blown hair strands, handheld composition, blur and grain around different actions, pursuing a rural life × old phone low-res × avant-garde grainy photo aesthetic. Generate one comprehensive preview image containing different actions for me to choose from.
This image was generated with GPT Image 2 by OpenAI. The prompt on this page is the exact text used to create it.
The example was generated with GPT Image 2 using: 9:16 · Cinematic · 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.