246사진 세 컷 편집 메뉴

Generate this image now. Return exactly ONE complete image directly. Do not ask questions, browse for photos, return a plan or give a prompt instead of the image.
Print exactly this heading: "【가게 이름】".
Print exactly these top-row words: "【메뉴 1】" and "【가격 1】".
Print exactly these middle-row words: "【메뉴 2】" and "【가격 2】".
Print exactly these bottom-row words: "【메뉴 3】" and "【가격 3】".
Those seven quoted values are the complete visible text. The photo composition sentences below are drawing directions only, never captions or subtitles. Each row has exactly TWO text groups, its menu name and its price, and no third line. Never print an English form key as the heading.

Output exactly one finished 2:3 portrait image, 1024x1536 pixels. Keep all important text inside the central 73% of image height: leave the top 12% and bottom 15% free of essential lettering.

[Input: values are the only facts and wording]
SHOP: "【가게 이름】"
MENU_1: "【메뉴 1】"
PRICE_1: "【가격 1】"
MENU_2: "【메뉴 2】"
PRICE_2: "【가격 2】"
MENU_3: "【메뉴 3】"
PRICE_3: "【가격 3】"

[INTERNAL PHOTO AND COLOR DIRECTIONS — NEVER PRINT ANY OF THESE VALUES]
DETAIL_1: "【메뉴 1의 사진 구성】"
DETAIL_2: "【메뉴 2의 사진 구성】"
DETAIL_3: "【메뉴 3의 사진 구성】"
TEXT_COLOR: "【글자 색】"
DISH_COLOR: "【접시 색】"

[Fixed card design]
Create one finished small restaurant menu page with THREE photographs and EXACT corresponding names and prices, in a restrained editorial magazine design. Warm off-white subtly fibrous paper, TEXT_COLOR lettering, thin rules and matte food photography. SHOP is the only header, above three equal-height horizontal rows. Each row's left approximately 55% holds one real-looking complete food serving in DISH_COLOR ceramic. Its right area holds only that exact MENU_n and PRICE_n. Align all three prices to the same right edge, with strong simple uncovered digits.
Bind every pair before drawing: top row MENU_1, PRICE_1, DETAIL_1; middle row MENU_2, PRICE_2, DETAIL_2; bottom row MENU_3, PRICE_3, DETAIL_3. DETAIL_n affects only its own photo and is not printed. When absent use that menu's ordinary food composition; no bonus side meal. Keep consistent soft natural light from the left, scale and restrained tone across the three photos, with toasted edges, moist cut surfaces and uneven vegetables. The photos are intentional three row frames within one page, not a 3x3 grid or a collage of outputs.
SHOP plus the three exact MENU names and their three PRICE strings are the entire visible text. All spaces, digits, punctuation and currency units remain unchanged. Never swap or recompute prices, invent old prices, discount, menu descriptions, English subtitles, row numbers, origin, QR or logo. Give the three rows enough height for names and numbers to be legible; long names wrap naturally inside their own row, without crossing into another. Keep the lower row as finished and readable as the first. This is one flat finished menu, no paper mockup. Newly imagined food photos must not be labeled proof of real portion size.

[Button meanings, internal instructions only]
TEXT_COLOR: "짙은 올리브" means deep olive; "짙은 남색" means deep navy; "코코아 갈색" means cocoa brown.
DISH_COLOR: "따뜻한 흰색" means warm white; "크림색" means cream; "옅은 회색" means pale gray.

[Missing inputs and finish]
An optional input line removed from this prompt is absent. Do not print its key, invent a substitute caption, ask for it or reserve an empty panel. Choose a fitting simple visual only where the fixed design explicitly permits an automatic default. Selected values alter content only; retain this card's single style and one composition. Every supplied visible string stays verbatim in its original script and currency, without translation. Ordinary automatic lettering, if explicitly allowed above, is Korean; internal English keys and button names are never captions.
Use the stated aspect ratio, adapting one coherent layout rather than stretching food or letters. Text-bearing posters keep important lettering within the central 16–83% of image height and 5–95% of width; extend food-related ornament, background, shadows or texture to the trim so these insets do not create empty bands. Keep all price digits and ingredient labels unobstructed. Text-free photographs fill the canvas naturally with their scene. No frame, phone, wall mockup, watermark or extra results.
No uploaded photo is needed: create a fictional food scene from these inputs. If a reference photo is attached, use it only for the stated food's visible shape and materials within this same drawing or photographic medium, without adding absent products or claiming exact portion preservation. Finish one image even when no photo is attached.

[Final output format]
Output exactly one finished 2:3 portrait image, 1024x1536 pixels. Keep all important text inside the central 73% of image height: leave the top 12% and bottom 15% free of essential lettering.
This selected output format takes priority over any fixed canvas size or aspect ratio elsewhere in this template. Adapt the composition, text zones and margins to the selected canvas without stretching or cropping essential content. Device safe areas apply only when a device is selected. Never print these format instructions on the image.