OpenAI has released ChatGPT Images 2.5, an enhanced version of its image generation technology. Following the second version introduced in April, this update aims to speed up image generation processes and improve quality. ChatGPT Images 2.5 offers sharper details and more precise editing capabilities. The company states that the new version reduces image generation latency by 50% compared to the previous version.

Improved Image Quality and New Features

The new version promises more natural lighting and richer textures. The technology is more successful at preserving subjects in reference photos and can follow editing instructions more reliably in multiple stages.

Users can now access a new Sketch feature that allows them to draw directly within ChatGPT. These drawings can be used as references in the visual creation process.

OpenAI is also introducing new templates for recognized image formats. Additionally, a new sharing system is being implemented so users can share their own visual creation ideas with others.

New API Options for Developers

The company also announced two different Images 2.5 versions that developers can use in their applications. GPT-Image-2.5 Flare and GPT-Image-2.5 Sunburst are available via the OpenAI API.

GPT-Image-2.5 Flare offers quality and speed improvements as a standard option. It caters to a wide range of use cases, from community content and artwork experiences to rapid visual prototyping.

GPT-Image-2.5 Sunburst is designed for premium image workflows requiring tighter control. This version is particularly preferred in creative processes requiring editing, such as campaign visuals or professional artwork photographs.

The update was released today for ChatGPT, ChatGPT Work, and Codex users on desktop, mobile, and web platforms. What do you think about these new features in ChatGPT Images 2.5?

OpenAI has introduced ChatGPT Images 2.5 technology to its users, offering sharper details and faster editing capabilities. Discover the new features.