# GPT Image Hub prompt

## Cảnh Đường Phố Phức Tạp Đa Hành Động
- URL: https://www.gptimagehub.com/vi/library/complex-multi-action-candid-street-scene
- Generator URL: https://www.gptimagehub.com/vi/generate?promptId=oss-ym-g2-31214
- Markdown URL: https://www.gptimagehub.com/api/agent/markdown?path=%2Fvi%2Flibrary%2Fcomplex-multi-action-candid-street-scene
- Category: Khác
- Model: gpt-image-2
- Aspect ratio: 16:9
- Style: 3d
- Tags: nhiếp ảnh đường phố, chụp lén, nhiều chủ thể, khung cảnh phức tạp, kiểm tra hiệu năng, kết xuất văn bản
- Updated: 2026-10-27T19:43:23.098Z
- Created: 2026-10-27T19:43:23.098Z

Tạo khung cảnh chụp lén chi tiết với nhiều hành động diễn ra đồng thời, văn bản rõ nét và bố cục tách bạch. Tùy chỉnh nhân vật nổi tiếng, con vật nền, logo và chữ trên biểu ngữ máy bay.

### Full prompt
```text
A RAW smartphone snapshot, completely unedited candid aesthetic, natural outdoor exposure, realistic daylight. All actions occur simultaneously inside the same frame with distinct spatial separation:
- Foreground center: [Will Smith] is accidentally tripping over a plate of spaghetti on the ground, tumbling forward in a comedic moment.
- Front-middle ground: Two distinct couples (four adults total) are engaged in a heated argument with expressive body language and gesturing hands.
- Far left: A man in a formal business suit has fallen onto the pavement, while an elderly woman in a green shirt bends down trying to help him stand up.
- Far right: Three people are dancing together wearing t-shirts clearly featuring the [OpenAI logo], each wearing a cap with a large readable number: 23, 44, and 98.
- Far background: A [black cat] sits on a windowsill looking outside through the glass.
- Sky: An airplane flies across the sky towing an advertising banner with crisp, readable text: "[@WolfRiccardo — NEW GPT-IMAGE MODEL COMING SOON!]".
Every character, action, number, and text string must be clearly discernible and sharply rendered.
```