OpenAI launches ChatGPT Images 2.5 with sketch-based creation and faster generation
OpenAI announced on September 8 that its ChatGPT Images 2.5 model now supports sketch drawing, image comments, and reduced generation latency.
OpenAI released an upgrade to ChatGPT Images 2.5 on September 8, introducing new capabilities including sketch-based creation and faster generation speeds. The company stated the update cuts generation latency by up to half compared with Images 2.0. New features allow users to draw rough visual references using a mobile keyboard shortcut instead of writing long text prompts. OpenAI also added image comments, letting users point directly at specific parts of an image to describe changes. The firm claimed this targets issues where editing one element unexpectedly alters surrounding subjects. The update maintains consistency across longer editing sessions while preserving people and pets from reference photographs. Currently, the model is available across ChatGPT, ChatGPT Work, and Codex on desktop, mobile, and web platforms.
