ChatGPT Images

ChatGPT Images is an integrated visual creation workspace within ChatGPT. Users can describe the image they want in a normal conversation, and the system generates it based on details such as the subject, composition, setting, lighting, style, text, and aspect ratio. The dedicated Images area at chatgpt.com/images also organizes generated work in one library, making it easier to revisit and reuse previous creations without searching through old conversations.

The platform supports a wide range of visual formats, including realistic photography, illustrations, posters, diagrams, advertisements, social media graphics, product mockups, logos, stickers, comics, and educational visuals. Its image model is designed to follow detailed instructions closely and can render written content inside images, making it useful for graphics where labels, headlines, signs, or interface text are important.

Users can also upload an existing image and describe the changes they need. ChatGPT can add, remove, replace, or transform objects, modify backgrounds, adjust styles, and preserve important details across successive edits. The built-in editor includes a selection tool for targeting a specific region, although users can also request broader changes directly through the conversation. Transparent-background creation and multiple aspect ratios are supported as well.

ChatGPT Images 2.0 is available across all ChatGPT tiers on the web, iOS, and Android. More advanced image creation with reasoning is available on selected paid plans, while generation limits and access can vary by subscription and demand. Developers can access related generation and editing capabilities through OpenAI’s image APIs, including GPT Image models and the Responses API.