OPENAI · Image model

GPT Image 2, built for practical image creation.

GPT Image 2 supports 1K output, common aspect ratios, and reference editing with 1 image per request for advertising, product visuals, and social content.

Text to image Reference editing

Create with GPT Image 2

CAPABILITIES

Built for practical image workflows.

Start from text

Describe the subject, material, light, and composition to create a finished visual.

Reference editing

Use up to 16 reference images to preserve a subject or visual direction.

Focused output

Generate one focused result per request for precise iteration.

OUTPUT SETTINGS

Quality: 1K
Common aspect ratios: auto · 1:1 · 4:3 · 3:4 · 3:2 · 2:3 · 16:9 · 9:16
Images per request: 1
Reference images: 16

PROMPT STARTER

Start with a clear visual direction.

Example prompt

“A structured launch poster for a design conference, bold geometric grid, accurate headline placement, black and electric pink palette.”