OpenAI Launches ChatGPT Images 2.5 with 50% Faster Image Generation and Improved Editing

Dynamic Beating AI News: OpenAI has officially released ChatGPT Images 2.5, with major upgrades in generation speed, human fidelity, and image 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 such as people and pets. The most practical improvement is that image editing is now significantly more responsive: users can directly circle the area they want to modify and write a simple instruction—such as “change only one petal” or “change the color of just one object”—without needing to repeatedly describe exact locations. When making multiple edits in sequence, the model is less likely to unintentionally alter areas not specified. ChatGPT now also supports sketching directly to control composition and includes common image templates such as posters and merchandise. Images 2.5 is now rolling out to ChatGPT, ChatGPT Work, and Codex users. The API has also launched GPT-Image-2.5 Flare and Sunburst. In the latest Arena benchmark, Sunburst ranks first and Flare ranks second, with both models collectively occupying the top two positions across all three categories: text-to-image, single-image editing, and multi-image editing.