OpenAI Announces ChatGPT Images 2.0

ChatGPT Images 2.0 is a more precise image generation capability for complex creative tasks of all kinds.

“OpenAI is launching ChatGPT Images 2.0 today, a new image generation model available across ChatGPT, Codex, and the API,” an OpenAI representative told me. “The update aims to make image generation more usable for real work, not just visual experimentation. Compared with prior versions, Images 2.0 is significantly better at following detailed instructions, rendering dense text, placing objects accurately, handling a wider range of aspect ratios, and generating images across more languages.”

ChatGPT Images 2.0 is ideally suited for small text, UI elements, diagrams, and dense layouts, OpenAI says. It supports aspect ratios from 3:1 to 1:3, so you can create assets for banners, slides, posters, and mobile formats. There are thinking-enabled workflows so ChatGPT can reason through tasks, access the web if needed, and generate multiple images from a single prompt. And you can request up to 8 outputs from a single request for making multiple formats of an image all at once.

In addition to its availability in ChatGPT and Codex, you can also access the underlying model, gpt-image-2, from the API so third-party developers can create their own solutions on top of these capabilities.

You can learn more about ChatGPT Images 2.0 on the OpenAI website.