OpenAI has released ChatGPT Images 2.5, a significant upgrade to its image generation capabilities. The new version produces images with more natural details, lighting, and textures while maintaining better fidelity to reference images and objects.
Performance improvements are substantial—the model generates images up to 50% faster than its predecessor. Beyond speed, the update brings enhanced precision to image editing. When you request specific changes, the model now alters only the elements you ask for while preserving the rest of the image intact, reducing unwanted modifications that plagued earlier versions.
The new model is immediately available to all users, making these improvements accessible without waitlists or staged rollouts.