GPT Image 2

image

OpenAI's image model for layouts, readable type, and precise edits. Attach up to four pictures and name them in the prompt.

Generate with GPT Image 2

What it supports

Aspect ratios: auto, 1:1, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 16:9, 9:16, 2:1, 1:2, 3:1, 1:3, 21:9, 9:21
Resolutions: 1K, 2K, 4K
Per request: up to 4

What it costs

1K: 4 credits
2K: 8 credits
4K: 12 credits

Credits come with a Generator plan. The exact cost is shown on the Generate button before anything is spent.

GPT Image 2 is OpenAI's image model for pictures that have to come out usable: a layout, type that reads, and an edit that keeps what you asked to keep. It is the successor to GPT Image 1.5, not a renamed older drawing model.

What it is good at

OpenAI describes GPT Image 2 as built for complex visual tasks, with stronger editing, better layouts, improved text rendering, and more reliable instruction-following. Use it when the prompt is a design, not a five-word mood.

How to use it on Visxy

Write the picture in the prompt. To edit or combine pictures, attach one to four of them and name them @image1, @image2, @image3, and @image4 in the prompt.

Where it differs

GPT Image 2.5 Flare is OpenAI's faster follow-on, and GPT Image 2.5 Sunburst is the one OpenAI describes as the extra step of precision. GPT Image 1.5 is the earlier model, and it is the one with a medium or high quality choice.

FAQ

Is there a separate editor?

No. Attach the pictures to this model and say what should change and what must stay.

How many pictures can I attach?

One to four, named @image1 onward.

Which newer model should I try?

OpenAI describes Flare as the fast everyday model and Sunburst as the one for tighter control. Start from the picture you need, not from the newest name.