OpenAI has introduced ChatGPT Images 2.5 with improved image editing capabilities.

OpenAI has released a new version of its image generation tool, ChatGPT Images 2.5. The update makes image creation faster, improves detail accuracy, and enhances element preservation during sequential editing.

Among the new features are more stable detail preservation across multiple edits and the ability to edit individual elements of an image using text comments without altering the rest of the work.

Users can also create sketches of their ideas directly in ChatGPT. To do this, simply enter the command Sketch. Additionally, the service now includes templates for popular formats, including posters and merchandise.

ChatGPT Images 2.5 begins rolling out on September 9 to ChatGPT, ChatGPT Work, and Codex users on desktop, mobile, and web platforms.

At the same time, OpenAI introduced two new API models. GPT-Image-2.5 Flare features improvements in generation quality, editing capabilities, and speed. GPT-Image-2.5 Sunburst is designed for more precise handling of detailed creative tasks, although its generation takes longer.

Source: OpenAI