OpenAI announced ChatGPT Images 2.5 on September 8, introducing a new feature called Sketch that lets users draw directly inside ChatGPT and use those drawings as prompts for AI image generation. Users can activate Sketch by typing @Sketch in the chat box, which opens a drawing window where they can sketch freely.
The feature transforms even crude drawings into detailed images. In a brief test, a simple doodle of a cat drawn with a computer mouse was successfully converted into a realistic photo based on text instructions. Users can also leave comments directly on parts of images they want changed — for example, requesting a cat's eyes be changed from blue to green.
ChatGPT Images 2.5 itself brings improvements to OpenAI's image generation capabilities. The company says the update produces images with more natural lighting and richer textures, and is better at following editing instructions across multiple turns of conversation. The update represents OpenAI's continued push to make its image generation tools more intuitive and accessible to casual users who may not be comfortable writing detailed text prompts.
The Sketch feature positions ChatGPT as a more visual and interactive creative tool, bridging the gap between imagination and AI-generated output in a way that feels natural and immediate.