Precise edits that preserve what matters, creative transformations that reimagine anything, and 4x faster generation. OpenAI's most capable image model, now available on Vofy.
Open VofyTry It
Set your parameters and start generating AI images with text, images references.
What's New
Precise edits that preserve composition and likeness, creative transformations from photo to any style, stronger instruction following, and denser text rendering — all 4x faster than the previous generation.
GPT Image 1.5 — Multi-Step Editing: Style Mix Output
When you ask for edits to an uploaded image, GPT Image 1.5 adheres to your intent down to the small details — changing only what you ask for while keeping lighting, composition, and people's appearance consistent. Add elements, subtract them, blend styles, or transpose subjects across scenes. You get the changes you want without losing what makes the image special.
GPT Image 1.5 — Golden Age Hollywood Movie Poster Transformation
GPT Image 1.5's creativity shines through transformations that change and add elements — like text and layout — to bring ideas to life while preserving important details. Turn a photo into a golden-age Hollywood movie poster, an 80s fitness instructor ad, a fashion campaign, or a painting. These transformations work for both simple and intricate concepts.
Before
AfterGPT Image 1.5 — Instruction Following: Previous Model
GPT Image 1.5 follows instructions more reliably than the previous version. Complex compositions with precise spatial relationships, specific counts, named elements, and multi-constraint prompts all come out as intended. The 6×6 grid example — 36 distinct objects, each correctly placed — shows the gap clearly.
Before
AfterGPT Image 1.5 — Text Rendering
GPT Image 1.5 takes another step ahead in text rendering, handling denser and smaller text than any previous version. Full newspaper layouts with headlines, body copy, and data tables come out legible. Infographics with labels, callouts, and structured content render accurately — making it practical for design mockups, editorial layouts, and branded content.
Try It NowSample outputs from GPT Image 1.5 — multi-step editing sequences, creative transformations, brand and e-commerce use cases. All images sourced from OpenAI's official release.
GPT Image 1.5 — Serene Seascape
Getting Started
Describe what you want to create, or upload an existing photo to edit. GPT Image 1.5 works from scratch or from reference — and accepts up to 16 input images for complex editing workflows.
Ask for precise edits that preserve what matters — lighting, composition, likeness. Or go further with creative transformations: turn a photo into a movie poster, a painting, or a fashion ad. The model changes only what you ask for.
Refine through multiple steps without starting over. At 4x the speed of GPT Image 1, you can run variations in parallel and build toward the final result quickly. Export as PNG, JPEG, or WebP.
Specifications
Evolution
| Feature | GPT Image 1 | GPT Image 1.5 |
|---|---|---|
| Generation Speed | Baseline | 4x faster |
| Editing Precision | Limited preservation | Keeps lighting, composition, likeness intact |
| Creative Transformations | Basic style changes | Movie posters, paintings, fashion ads, ornaments |
| Instruction Following | Approximate | Precise — spatial relationships, counts, named elements |
| Text Rendering | Inconsistent | Dense — newspapers, infographics, mockups |
| Pricing (image output) | $40/1M tokens | $32/1M tokens (20% cheaper) |
Use Cases
From social content to professional workflows, see how creators and teams are using AI image generation across industries.
Generate campaign visuals, product lifestyle shots, and branded graphics with consistent logo placement and accurate color reproduction. GPT Image 1.5 is well suited for marketing teams generating full image catalogs — variants, scenes, and angles — from a single source image.
Produce product images across multiple backgrounds and lighting setups from a single reference photo. Precision editing preserves fine product details — stitching, reflections, label text — while swapping environments. Generate full catalogs at 4x the speed of the previous model.
From scientific posters and editorial illustrations to concept art and publication covers, GPT Image 1.5 handles complex compositions with accurate detail. Dense text, structured layouts, and intricate visual relationships all render correctly.
Transform photos into movie posters, paintings, fashion ads, or any visual style — while preserving the subject's likeness and key details. Multi-step editing lets you iterate through a sequence of changes without starting over, building toward the final result.
Everything you need to know about GPT Image 1.5.
Generate, edit, and iterate on images with the most instruction-accurate model available — text rendering, precision editing, and 4x speed, all in one studio.
Open VofyGPT Image 1.5 is developed by OpenAI. Product information is sourced from OpenAI's official documentation and announcements.