GPT Image 2
Prompts
OpenAI's most advanced image generation model. These prompts are crafted and optimized specifically for it. Copy, paste, generate.
What is GPT Image 2?
GPT Image 2 is OpenAI's most advanced image generation model, released in April 2025 as a native capability inside ChatGPT. Unlike standalone tools like DALL-E or Midjourney, GPT Image 2 works conversationally — you describe what you want, get an image, then refine it through follow-up messages.
It excels in three areas: accurate text rendering (perfect for posters, UI mockups, and signage), photorealism (images nearly indistinguishable from real photographs), and complex instruction following (it handles multi-part prompts with precision).
All GPT Image 2 Prompts
23 prompts across 8 categories. Filter by type, copy any prompt, or view the full breakdown.
AI Street Photography Prompt — Rainy Tokyo Night
A candid street photograph taken in Tokyo on a rainy evening. A lone figure with a transparent umbrella walks…
AI Macro Photography Prompt — Eye with Four Seasons
An extreme macro photograph of a human eye, shot at 1:1 magnification. The iris contains an impossibly detaile…
AI Photography Prompt — Japanese Convenience Store at Night
A photorealistic image of a Japanese convenience store (konbini) at 2am. Warm fluorescent light spills out thr…
AI Portrait Prompt — Golden Hour Natural Photography
A close-up portrait of a young woman with freckles, shot during golden hour. Warm sunlight catches individual…
AI Food Photography Prompt — Professional Ramen Shot
Professional food photography of a bowl of tonkotsu ramen. Rich, creamy white broth with a thin layer of pork…
Studio Ghibli Style AI Prompt — Photo to Anime Transformation
Transform this photograph into a Studio Ghibli anime style illustration. Keep the exact same composition, pose…
AI Background Replacement Prompt — Fantasy Landscape
Keep the person in the foreground exactly as they are — same pose, clothing, lighting on their body, and facia…
Vintage Film Color Grading AI Prompt — 1970s Kodachrome
Apply a vintage 1970s film color grade to this photo. Shift shadows toward teal/cyan, highlights toward warm a…
AI Pet Photo Editing Prompt — Dress Your Pet in Outfits
Take this photo of my cat/dog and dress them in a tiny, perfectly tailored business suit — complete with a min…
AI Character Expression Sheet Prompt — 16 Emotions Grid
Create a 4x4 grid character expression sheet for an anime-style girl with short blue hair, large amber eyes, a…
AI App UI Design Prompt — Dark Mode Music Player
Design a mobile music player app screen (iPhone 15 Pro dimensions, 393x852px). Dark mode UI with a deep navy (…
Film Noir AI Art Prompt — Cinematic Detective Scene
A cinematic film noir scene. A lone detective in a trench coat and fedora stands in a rain-soaked alley at nig…
AI Product Photography Prompt — Luxury Perfume Shot
A luxury perfume bottle product photography shot. The bottle is made of deep amber glass with a gold geometric…
AI Infographic Prompt — Coffee Brewing Methods Guide
A beautiful vertical infographic about coffee brewing methods. Title: 'The Art of Coffee Brewing' in elegant s…
AI Manga Colorization Prompt — Black & White to Color
Take this black and white manga panel and colorize it with a vibrant anime color palette. The character has: b…
AI Christmas Photo Prompt — Family Portrait with Matching Pajamas
A warm, photorealistic Christmas family portrait. A family of four (parents in their 30s, a boy around 7, a gi…
AI Game Remaster Prompt — Classic to Next-Gen Graphics
Recreate this classic video game screenshot as a next-generation AAA title, as if the game was remade in 2025…
AI Selfie Prompt — Photorealistic Lifestyle Portrait
Generate a photorealistic selfie portrait using the following structured parameters. Scene: bedroom interior,…
AI 90s Retro Prompt — Video Game Character Candid Photo
A grungy analog photograph of Mario (the Nintendo character) sitting on the floor of a dimly lit 1990s bedroom…
AI Action Figure Prompt — Custom Toy in Blister Pack
Create a stylized action figure of a person in a premium collectible toy blister pack. The figure should be ap…
AI 3D Icon Design Prompt — Fluffy Plush App Icons
A set of 9 app icons arranged in a 3x3 grid, each rendered as a 3D fluffy/plush object on a clean pastel backg…
Pixar Style AI Prompt — 3D Cartoon Avatar from Photo
Transform this photo into a Pixar-style 3D animated character. The character should have: slightly exaggerated…
AI Oil Painting Prompt — Dutch Masters Style Transfer
Transform this photograph into a classic oil painting in the style of the Dutch Golden Age masters (Vermeer, R…
Tips for better prompts
Get better results by following these proven techniques for GPT Image 2.
Use Hex Color Codes
GPT Image 2 understands hex colors precisely. Instead of saying "blue background", say "deep navy background (#1a1a2e)". This gives you exact control over the color palette.
Specify Camera Settings
For photorealistic results, include camera details: lens ("85mm f/1.4"), film stock ("Kodak Portra 400"), and shooting conditions ("golden hour, backlit").
Define Aspect Ratios
Always specify your desired aspect ratio ("16:9", "3:4", "1:1") and resolution ("4K"). This prevents awkward cropping and ensures your output fits the intended use.
Describe Lighting Explicitly
Lighting makes or breaks an image. Be specific: "warm directional sunlight from the upper left casting long shadows" beats "good lighting".
Leverage Text Rendering
GPT Image 2’s standout feature is accurate text in images. Use it for posters, UI mockups, book covers, and signage. Wrap text in quotation marks.
Iterate Conversationally
Unlike other generators, GPT Image 2 lets you refine images through conversation. Start with a detailed prompt, then follow up with specific changes.
GPT Image 2 FAQ
What is GPT Image 2?
GPT Image 2 is OpenAI’s latest image generation model, released in April 2025. It’s built natively into ChatGPT and represents a major leap in AI image generation — excelling at photorealism, accurate text rendering, precise instruction following, and maintaining consistency across multiple edits to the same image.
How do I use GPT Image 2 prompts?
Simply copy any prompt from this page, open ChatGPT (with GPT-4o or later), paste the prompt, and hit send. GPT Image 2 is the default image model in ChatGPT — no special setup needed. For best results, use the prompt exactly as written, then iterate with follow-up instructions.
Are these GPT Image 2 prompts free to use?
Yes, every prompt on this page is completely free. No sign-up, no paywall. Copy any prompt and use it in ChatGPT. We add new GPT Image 2 prompts regularly across all categories.
What makes GPT Image 2 different from DALL-E or Midjourney?
GPT Image 2 stands out in three key areas: it renders text inside images with near-perfect accuracy, it follows complex multi-part instructions more faithfully than other models, and it works conversationally — you can ask ChatGPT to edit, adjust, or iterate on the generated image in follow-up messages without starting over.
Can I use these prompts in other AI image generators?
While these prompts are optimized for GPT Image 2, most work well in Midjourney, DALL-E 3, and other generators. The core descriptions translate across models. You may need to adjust model-specific parameters (like Midjourney’s --ar flag for aspect ratios), but the creative direction remains the same.
How do I write better GPT Image 2 prompts?
The best GPT Image 2 prompts are specific and structured. Include: a clear subject, art style or medium, lighting conditions, camera angle or composition, color palette (hex codes work great), and output specifications like aspect ratio and resolution. Check our tips section above for detailed guidance.