OpenAI has begun integrating animated GIF creation directly into ChatGPT, a capability driven by its yet‑unannounced Images 2.6 model. According to Decrypt, users will be able to describe an animation in plain English or upload a photo and have ChatGPT generate a looping GIF, complete with controls for playback speed, frame‑by‑frame inspection, and editing. The rollout is incremental, meaning the feature is currently limited to a subset of accounts.

How the GIF maker works

The new tool leverages Images 2.6, the latest iteration of OpenAI’s image‑generation engine, which follows the September release of Images 2.5. While Images 2.5 improved detail and speed, Images 2.6 adds motion synthesis by producing a sequence of frames that are stitched into an animated GIF. In demonstrations, a simple prompt such as “make a gif of a cat dancing” yielded a short animation that users could fine‑tune via on‑screen sliders. The interface also allows inspection of individual frames, enabling precise adjustments before sharing.

Developer access and API extensions

OpenAI is extending the Images 2.6 capabilities to developers through two new API models: GPT‑Image‑2.6 Flare, optimized for faster generation, and GPT‑Image‑2.6 Sunburst, aimed at higher‑quality outputs. These APIs will be available to both ChatGPT and Codex users across web, desktop, and mobile platforms, according to The Verge’s reporting cited by Decrypt. The addition of motion generation follows OpenAI’s March shutdown of Sora, its standalone AI video‑generation app, suggesting a strategic shift toward embedding animation features within existing products rather than maintaining separate services.

What to watch next

OpenAI has not disclosed a timeline for universal availability, so monitoring the gradual rollout will be key. Future updates may expand GIF length, resolution, or introduce more complex motion beyond simple loops. Additionally, the performance of the Flare and Sunburst APIs will likely be benchmarked against competing models such as Google’s Nano Banana 2, as was done with Images 2.5. Finally, the broader impact on content creation workflows and potential integration with other OpenAI tools will be an area of interest for developers and creators alike.