OpenAI has launched a new image generation model, ChatGPT Images.

OpenAI announced the launch of a new image generation model, ChatGPT Images. The new model is based on an improved flagship system and is four times faster than previous versions.

Dec 25, 2025 0 2

OpenAI has launched a new image generation model, ChatGPT Images.

Now, users can create images from scratch or edit uploaded photos and get results that precisely match their needs. The model makes detailed edits while preserving key elements of the original, such as lighting, composition, and human likeness. This allows it to be used for practical photo editing, virtual try-on of clothes and hairstyles, and stylistic and conceptual transformations.

ChatGPT Images is designed to make the image generation process simpler and more intuitive. Users can either formulate queries in text or choose pre-made styles and ideas without having to write detailed instructions.

The model efficiently performs various types of editing, including adding, removing, combining, and transforming elements, including text and layout. It also preserves the unique features of the original image. Particular attention is paid to improved rendering of text and details such as small faces, increasing the naturalness and usability of the result.

ChatGPT Images is available to all ChatGPT users and is available to developers through an API called GPT-Image-1.5. This feature is available to most users, with enterprise customers expected to receive access later.