Tools & services

ChatGPT Images

People & organizations
OpenAI

ChatGPT Images creates and edits images inside the ChatGPT conversation, so you describe the result and the revisions in words.

Images inside the conversation

ChatGPT Images generates and edits images within ChatGPT. Ask for a study at dusk in watercolor and it produces one. Upload an image of your own and you can ask it to change only the background, or remove an object.

Why working in chat helps

You do not have to get everything into one prompt. Looking at a result, you can say darker, no people, make it landscape and keep going. Because it uses the previous image and the conversation as context, it is easy to iterate even without experience in design software.

  • Generate a new image from text
  • Edit an uploaded photograph or illustration
  • Place text and fine details within an image
  • Choose an aspect ratio and make the background transparent
  • Revise only a selected region

What it suits

Everyday production: blog featured images, social posts, illustrations for slides, product concept images, removing unwanted objects from photographs. Being combined with ChatGPT’s text understanding also means you can hand it an article and ask for a single image that represents the topic.

To enlarge a small image, editing and upscaling are sometimes combined. Note that a regenerated image is not identical to the original — details shift.

Cautions

Text, hands, and small logos are frequently wrong. Inspect at full size before publishing. Requests that resemble real people or copyrighted work raise personality rights, copyright, and terms-of-service considerations.

Model names, usage limits, and plan availability change. When writing about it or advising others, check OpenAI’s current help documentation.

Related terms

Sources and review information

Last reviewed July 17, 2026

Back to the AI Glossary

Search this site