**[ChatGPT Images 2.5 Launched: 50% Faster Image Generation, Precise Editing Now Possible]** Beating AI Newsflash: OpenAI has officially released ChatGPT Images 2.5, focusing on upgrades in generation speed, character fidelity, and editing experience. Compared to Images 2.0, the new version reduces image generation latency by up to 50%. When using photos as references, it better preserves the appearance and details of subjects like people and pets. The most practical improvement this time is that editing has become much more "obedient." Users can directly highlight the area they want to modify on the image and provide a simple instruction, such as changing just one petal or altering the color of a single object, without repeatedly describing the exact location in text. When making multiple rounds of edits, the model is also less likely to change areas that were not specified for modification. ChatGPT now supports sketching directly to control composition and has added commonly used image templates for posters, merchandise, and more. Images 2.5 is now available to ChatGPT, ChatGPT Work, and Codex users. The API has also launched GPT-Image-2.5 Flare and Sunburst. In the latest Arena evaluations, Sunburst ranked first and Flare second, with both models dominating the top two spots across three leaderboards: text-to-image generation, single-image editing, and multi-image editing. [Original Link]