Bilingual Prompt Comparison Gallery — промпт для GPT Image 2

Загрузка...

Bilingual Prompt Comparison Gallery

ИнфографикаGPT Image 22 поля можно заменить

Objective: Create a widescreen screenshot-style comparison panel showcasing the image results generated by the same cue word in two languages. Canvas: Use a 1199×461 pixel horizontal composition with a very light gray or off-white background, similar to a social media screenshot or a web showcase. Layout: The left 75% area features a 2x4 showcase containing 8 rounded-corner photo cards with narrow white gaps between them. Each card is a nearly square portrait of two women sitting together, facing each other or leaning inwards, each placing an index finger to the other's lips in a shared "shh" gesture. The right 25% area contains two stacked cue word description blocks, aligned with the two rows of the showcase. Showcase Details: Contains 8 different generated photo cards: 1. Top Left: Two women sitting in front of a blue studio background, one wearing a floral dress and the other an orange outfit. 2. Second from the top row: Two women in casual sweaters, one with dark hair and the other with blonde hair, sitting together against a light gray background. 3. Top row, third: Warm, cinematic cyan/orange tones image; two women wearing hats, faces obscured by large, softly blurred rectangles. 4. Top row, fourth: Two women against a white background; one in white, the other in a mustard yellow cardigan with a striped shirt. 5. Bottom left: Two young women seated by an indoor window in soft, casual snapshot lighting. 6. Bottom row, second: Two young women in black and light pink tops in a casual close-up indoor portrait. 7. Bottom row, third: Two young women seated in chairs; one in a dark dress, the other in a light-colored cardigan, against a plain wall background. 8. Bottom row, fourth: Two young women in white and gray sweaters seated in front of a light-colored indoor wall. Face Treatment: All visible faces on all 8 cards should be anonymized by covering the facial areas with soft rectangular mosaic/blur blocks to match a screenshot-like privacy masking effect. Right-side Text: Display English prompt text in the top right corner: Two women are sitting facing each other. They are each placing their index finger on the other's lips. Below this are two small, round, gray labels that read "raw" and "v 8.1" respectively. Below that, aligned with the second line, the Japanese prompt text is displayed: Two women are sitting facing each other, each with their index finger on the other's lips. Below this, place two identical gray labels, one reading "raw" and the other "v 8.1". Visual Style: Photorealistic generated image thumbnails, varied lighting and clothing, clean UI spacing, rounded corners, dark gray sans-serif interface text, and small light gray capsule labels. The overall image should resemble a screenshot of a consistent contrast grid, rather than a meticulously retouched poster. Constraints: Use 8 photo cards, 2 text blocks, and a total of 4 capsule labels. Do not add logos, watermarks, browser borders, descriptive text within photo cards, or additional UI elements.

GPT Image 2

Что можно заменить

Подсвеченные слова — поля шаблона. Кликните по любому и впишите своё: бренд, цвет, персонажа. Остальной текст отвечает за свет и композицию. «В студию» откроет генерацию с этим промптом, «Скопировать» положит его в буфер — уже с вашими правками.

Похожие промпты