OpenAI launches ChatGPT Images 2.6 with animated GIF creation

OpenAI shipped ChatGPT Images 2.6 on November 4, adding native animated-GIF generation and persistent reference images across ChatGPT and Codex clients on web, desktop and mobile. The headline feature is animated GIFs, generated in a single step. Users can generate them from either a text prompt or an uploaded photo, and the model returns a looping animation. Users also get controls once the GIF exists. They can adjust playback speed and edit individual frames. OpenAI pitches the feature for animated mascots, social reactions, product loops and storyboards used to explore movement before video production.

For developers, OpenAI is adding two new models, GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst. Flare is tuned for speed, while Sunburst targets higher quality output. Images 2.6 generates images 50% faster than before, demonstrating significant improvements in performance across the board. Images are already one of ChatGPT's highest-volume uses, and OpenAI says that volume reached 4 billion images per week across ChatGPT, Codex and the API. It is rolling out to all ChatGPT and Codex users on the web, desktop and mobile.