OpenAI has unveiled a new image generation model and added an “Images” section to ChatGPT. The new model, called GPT Image 1.5, operates up to four times faster than its previous version. Rolling out gradually to all users, the system understands prompts more accurately, renders even small text on images without errors, and produces highly detailed visuals.
The new model also comes with significantly improved editing capabilities. Users can now easily add new elements to existing images, remove objects, modify visuals, or combine multiple images into a single scene. For example, several different photos can be uploaded and merged into one cohesive composition.
There have also been updates to the app’s interface. The former “Library” section has been replaced by “ChatGPT Images.” In this section, users can view previously generated images and receive ready-made ideas and style suggestions for new creations. After describing the desired image, the generation process continues seamlessly within the chat window, just as before.
