OpenAI has built a GIF generator directly into ChatGPT, allowing users to turn simple text descriptions or uploaded photos into smooth looping animations without opening standalone editing software or manually stitching together image frames. The feature is powered by an undisclosed AI model called Images2.6 and supports adjusting playback speed, viewing animations frame by frame, and making quick edits without leaving the conversation window.
OpenAI is gradually rolling out the Images2.6 model to ChatGPT and Codex users across web, desktop, and mobile apps. The release comes months after the company shut down its standalone Sora AI video-generation app in March of this year. Although looping GIF animations are simpler than full AI video projects, integrating dynamic generation capabilities directly into the main chat interface significantly lowers the barrier to creating simple animations. In recent weeks, OpenAI also introduced a sticker-making tool in ChatGPT.
As part of the Images2.6 rollout, developers are also getting dedicated API tools: OpenAI released GPT-Image-2.6Flare for speed-oriented tasks and GPT-Image-2.6Sunburst for high-visual-quality projects. These two developer models directly continue the dual-release pattern introduced with the Images2.5 version last September. The total number of images created across OpenAI's platforms continues to rise. The company says users now create more than 4 billion images per week through ChatGPT and its developer API, a sharp jump from the 3 billion per week announced in September.
The new native GIF creation feature makes ChatGPT a more flexible tool for everyday visual content creation. With frame-by-frame editing controls built into the chat window, users can quickly create meme loops without third-party apps. This integration reflects how AI image generation is converging from standalone tools into the main conversation interface, further lowering the barrier to dynamic content creation.
