Skip to content

OpenAI reportedly rolls out GIF creation to ChatGPT and Codex users

The Images 2.6 rollout covers web, desktop and mobile apps, with users able to animate a text prompt or photograph and edit the resulting GIF.

A woman uses a laptop displaying a red-and-blue wave image and a slider; a picture of an orange flower rests beside it. SI illustration
In this article

OpenAI is rolling out its Images 2.6 model to all ChatGPT and Codex users on the web and in desktop and mobile apps, bringing animated-GIF generation, The Decoder said, citing The Verge. Users can make a GIF from a text prompt or an uploaded photo, then adjust its speed or edit individual frames.

The update follows Image 2.5, which introduced the Flare and Sunburst variants. Images 2.6 is also intended to speed up image generation overall.

A dancing cat and developer models

In an OpenAI demonstration, a prompt for a dancing cat produced slightly different images of a character, assembled like a flipbook into a GIF. The Decoder assessed that the character appeared recognizable across frames; it suggested earlier models would likely have changed its appearance too much for the approach to work.

OpenAI is also releasing GPT-Image-2.6 Flare and Sunburst for developers through its application programming interface, or API, which lets software connect to the models. Flare is intended for efficient image generation and Sunburst for higher-quality images.

According to OpenAI, users now generate well over four billion images a week through ChatGPT and the API, one billion more than in September, The Decoder reported.

All Technology stories