awesome-gpt-image-2-5-prompts

Overview of awesome-gpt-image-2-5-prompts

What it is

A curated, credited library of 150 prompts for OpenAI's GPT Image models, including GPT Image 2.5. Every prompt is reproduced verbatim from the person who published it, with their account and a link to the post it came from. The library contains 150 prompts from 110 creators across 9 use cases, with 25 written as JSON. It was updated on 2026-09-10. The prompts are natural-language and not model-specific, so a prompt written for GPT Image 2 runs unchanged on 2.5. All numbers in the file are computed from data/prompts.json when the file is built.

Key features

• 150 prompts reproduced verbatim from their authors, with credit and links to the original posts.
• 110 creators credited.
• 9 use cases: Posters, Text in image, Infographics, UI mockups, Product shots, Ad creative, Character design, Portraits, Illustration.
• 25 prompts written as JSON.
• 56 prompts contain template tokens in curly braces; 18 use bracketed capitals such as [BRAND NAME].
• Numbers (prompts, creators, use cases, characters) are computed automatically from data/prompts.json.
• Prompts are natural-language and not model-specific, so they work on GPT Image 2.5 unchanged.

Use cases

The prompts are grouped by what you are trying to make:

• Posters – Movie teasers, event flyers, travel posters and album covers. Focus on layout and composition.
• Text in image – Legible words rendered inside the picture, with exact quoted strings and typeface specifications.
• Infographics – Charts, timelines, exploded views and explainer slides; many written as JSON.
• UI mockups – Dashboards, mobile screens, landing pages and stream overlays; components must be named.
• Product shots – Studio product shots, packaging renders and catalog imagery with precise lighting and camera positions.
• Ad creative – Campaign key visuals, banners and social ad creative with both image direction and copy.
• Character design – Turnaround sheets, expression grids, equipment breakdowns and mascots; often asking for a sheet rather than a single picture.
• Portraits – Editorial portraits, headshots and film-look photography; rely on camera direction.
• Illustration – Anime, watercolour, line art, 3D render, isometric and pixel art; style prompts are the shortest.

Getting started

See the repository README for instructions on how to use the prompts. The prompts are available in the prompts/ directory (one file per use case) and in data/prompts.json.

License and pricing

The repository metadata indicates a license of 'NOASSERTION' and is listed as open source. The README does not state a specific license. The project is free to run.

Based on the project's public README and repository metadata.