Download app
Download app

GPT Image 1.5

GPT Image 1.5 is a precision-focused image generation and editing model built for layout accuracy, spatial reasoning, and clean text rendering. In Recraft, it behaves more like a controllable visual editor than a random image generator.
Up to 30 free image generations per day
Trusted by designers at

How to use GPT Image 1.5

1
Select or import your image(s) onto the Recraft canvas, or generate your own within Recraft
2
Select Model → GPT Image 1.5 and describe layout, edits, or placement in natural language
3
Generate consistent results and export in PNG, JPG, TIFF, PDF, or SVG

Layout-aware, text-accurate generation

GPT Image 1.5 prioritizes spatial accuracy and deterministic layout control over stylistic randomness. It understands grids, coordinates, and descriptive placement, making it effective for UI elements, HUDs, sprite sheets, diagrams, and layout-heavy designs. Its strength lies in localized edits and consistency locking, enabling repeatable visual identities across production pipelines.

FAQ

What is GPT Image 1.5 best suited for?

UI layouts, game assets, diagrams, and designs that require spatial precision.

Does it support localized edits?

Yes. It can modify specific regions without disrupting the overall layout.

How does it handle text?

It renders text clearly and places it accurately within compositions, with reliable placement for layout-heavy designs.

For each GPT Image 1.5 Low image, how many credits are used and how long does it take?

Each image costs 1 credit and takes about 15 seconds.

For each GPT Image 1.5 Medium image, how many credits are used and how long does it take?

Each image costs 5 credits and takes about 25 seconds.

For each GPT Image 1.5 High image, how many credits are used and how long does it take?

Each image costs 18 credits and takes about 50 seconds.