ChatGPT is gaining the ability to generate animated GIFs natively, from either text prompts or uploaded photos, according to a report by The Verge. The feature runs on Images 2.6, a new OpenAI image-generation model that has yet to receive a standalone official announcement, and the rollout is gradual—most users cannot access it yet.

Users can describe an animation in plain English or upload an existing photo and ask ChatGPT to bring it to life. Once the GIF is generated, they can adjust its playback speed, inspect individual frames, make edits, and share the finished animation. GIFs have been around since 1987, and the format's short, looping animations remain a staple of social media and messaging apps, so the feature brings a familiar web format into the same place users already write prompts and edit images.

OpenAI demonstrated the capability with a simple request: “can you make a gif of a cat dancing?” The resulting interface reportedly includes controls to change the animation speed and examine the GIF frame by frame.

Decrypt tested the feature, but it was not available on its accounts at the time of testing. In that interaction, the model produced a series of frames rather than the animation itself. With the rollout appearing to be gradual, users may have to keep asking for their dancing cats a little longer.

Images 2.6 Is Also Coming

The GIF maker is reportedly part of a broader Images 2.6 update that promises faster image generation. Per The Verge, OpenAI is bringing the model to ChatGPT and Codex users across web, desktop, and mobile platforms. Speed matters more than usual here, because a GIF packs many individual frames into a single animation—precisely the kind of workload where generation time adds up.

Developers are also getting two new API models: GPT-Image-2.6 Flare, designed for faster generation, and GPT-Image-2.6 Sunburst, aimed at higher-quality results. OpenAI has not yet published a standalone announcement detailing the new model's capabilities, and its performance could not be independently verified.

The update follows the September release of ChatGPT Images 2.5, which introduced sharper details, more precise editing, and generation speeds up to 50% faster than its predecessor. In Decrypt's previous testing, Images 2.5 went head-to-head with Google's Nano Banana 2 across six categories, with each model winning three. OpenAI's model delivered particularly impressive illustration quality and improved on the oversharpening problems that affected earlier versions, although it still made mistakes when rendering text.

The move into GIF generation also comes months after OpenAI shut down Sora, its standalone AI video-generation app, in March. Animated GIFs are considerably simpler than the videos Sora generated, but the new feature brings a form of motion generation directly into ChatGPT without requiring a separate application—placing text, images, and now short looping animation inside a single assistant.

OpenAI has not provided a specific date for when every ChatGPT account will receive GIF generation. Until then, the gradual rollout and the still-absent standalone announcement for Images 2.6 are the details worth watching, since either would clarify what the new model can actually do and who can use it.