Anime School Rom-Com Key Visual

Generates a wide anime classroom promotional scene with three students, Japanese title overlays, and detailed rom-com character dynamics.

Prompt by @MirochillView the original post โ†—Published 2026-08-19CC BY 4.0

The prompt

Fill in the blanks โ€” 2 placeholders

Edit any of these and the prompt below updates. Copy gives you the filled-in version.

Goal: Create a cinematic anime key visual for a school rom-com scene titled ่ฒ ใ‘ใƒ’ใƒญใ‚คใƒณใŒๅคšใ™ใŽใ‚‹!, with polished modern TV-anime rendering, natural lighting, and coherent character designs.

Canvas: 16:9 horizontal frame, 1200ร—675 composition, set inside a bright but slightly shadowed school classroom or club room with large windows in the background showing blue sky, clouds, railings, and distant buildings. Use soft daylight from the windows, cool gray concrete pillars, shallow depth of field, and subtle filmic contrast with no visible grain.

Layout: Three high-school students sit at desks in the foreground and midground. Place the main girl large on the left, the boy centered slightly behind her, and the second girl on the right, smaller and partially separated by a pillar. Add anime-style promotional text overlays and handwritten Japanese comments around the characters.

Characters: Exactly 3 characters. 1) Main heroine on the left: a cheerful teenage girl with deep blue shoulder-length messy hair, blue eyes, blushing cheeks, and a teasing smile; she wears a white short-sleeve school shirt with a blue bow tie and a cute decorative chest accessory, resting her cheek on one hand as she leans forward confidently. 2) Center boy: a nervous teenage boy with short dark brown hair, white short-sleeve school shirt, dark green necktie, and a slightly sweaty awkward expression, looking toward the main heroine as if trapped in a conversation. 3) Right-side heroine: a quiet teenage girl with short burgundy-red bobbed hair and hairpins, wearing a white school shirt, sitting lower behind a dark school bag while hiding behind or looking down at a small smartphone, appearing jealous or indifferent.

Foreground objects: Include exactly 4 prominent desk items: a pink rectangular lunch box or pouch in front of the blue-haired girl, a crumpled pale pink cloth beside it, a small strawberry yogurt drink carton with a straw in front of the boy, and a large dark navy school bag across the right foreground.

Text content: Include exactly 7 visible text elements. 1) Top-left title: ใ€Œ่ฒ ใ‘ใƒ’ใƒญใ‚คใƒณใŒๅคšใ™ใŽใ‚‹!ใ€ with small English subtitle โ€œMAKEINE Too Many Losing Heroines!โ€ beneath it. 2) Handwritten comment near the main heroine on the left: ใ€Œใ‚„ใฃใฑ ็งใฏ่ฒ ใ‘ใƒ’ใƒญใ‚คใƒณใชใ‚“ใ ใ‚ˆใญใ€œใ€ with a curved arrow pointing to her. 3) Small thought text near the boy: ใ€ŒใพใŸ ๅง‹ใพใฃใŸโ€ฆใ€. 4) Poster on the center-right wall: ใ€Œ่ฒ ใ‘ใฆใ‚‚ใ€ใใฃใจใ€้’ๆ˜ฅใฏใ€็ต‚ใ‚ใ‚‰ใชใ„ใ€‚ใ€ with small English โ€œToo Many Losing Heroines!โ€ beneath. 5) Handwritten quote in the upper-right corner: ใ€Œใใ‚Œใงใ‚‚ใ€ๅฅฝใใชๆฐ—ๆŒใกใฏใ€ใใฃใจใ€่ฒ ใ‘ใ˜ใ‚ƒใชใ„ใ€‚ใ€ with a thin blue underline. 6) Handwritten comment near the red-haired girl: ใ€Œๅˆฅใซ ๆฐ—ใซใ—ใฆใชใ„ใ€ with a dotted line and arrow pointing to her. 7) On the drink carton, write ใ€Œใƒจใƒผใ‚ฐใƒซใƒˆ ใ„ใกใ”ใ€ with a strawberry graphic.

Visual style: High-quality Japanese anime production still, delicate line art, expressive faces, soft cel shading, realistic classroom perspective, slightly muted colors, natural blue daylight, detailed hair highlights, emotional rom-com atmosphere, promotional poster composition.

Constraints: Keep the composition faithful to a single wide anime screenshot, do not add extra characters, do not add extra desk objects beyond the four prominent items, keep all Japanese text legible and placed as described, avoid photorealism, avoid 3D rendering, avoid watermarks.

Generate opens this prompt in the editor, ready to run.

Model
GPT Image 2
Media
Image
Language
English
Words
510
YouTube ThumbnailAnime / MangaSketch / Line ArtCinematic / Film StillFood / DrinkLandscape / NatureCharacter

How to use this prompt

  1. 1Copy the prompt โ€” fill in any placeholders first and the copy takes your version.
  2. 2Paste it into your image model, or press Generate to open it here with your own photo.
  3. 3Run it twice. Generation is not deterministic and the second pass is often the keeper.
  4. 4Finish it with the free tools: upscale, crop to the platform you need, compress before sharing.

Go beyond generating the image

Everything after the generation, on the same site.

Upscale to 4K โ†’Remove the background โ†’Crop for a platform โ†’Compress before sharing โ†’All AI apps โ†’

More from GPT Image 2

Same model, same dataset โ€” GPT Image 2 prompts people have shared.

View all GPT Image 2 prompts โ†’

All image categories

Use cases
Styles
Subjects