OpenAI 為 ChatGPT 推出動畫 GIF 功能及 Images 2.6 圖像模型

OpenAI is adding animated GIFs to ChatGPT with a new image model, Images 2.6. Users can make a GIF from text or a photo and edit its frames. Developers get two new API models, and OpenAI says people now make over 4 billion images a week.

OpenAI is rolling out animated GIFs in ChatGPT alongside a new image model, Images 2.6, The Verge reported on Wednesday. Users can turn a written prompt or an uploaded photo into a GIF.

Once a GIF is made, users can change how fast it plays, look at and edit single frames, and share it. Until now, ChatGPT could not make a GIF in one step. The feature arrives months after OpenAI shut down Sora, its AI video app, in March.

OpenAI says Images 2.6 also makes image generation in ChatGPT faster. It is rolling out to all ChatGPT and Codex users on the web, desktop and mobile.

Developers get two new API models: GPT-Image-2.6 Flare for speed and GPT-Image-2.6 Sunburst for quality. They follow the split OpenAI introduced with Images 2.5 in September, when Adobe put one of its models into Firefly.

Use keeps growing. OpenAI says people now make more than 4 billion images a week across ChatGPT and its API, up from 3 billion in September. The figures are OpenAI’s own and cover all image use, not only GIFs.

OpenAI has updated its image tools often this year. Images 2.0, launched in April, was its first image model that reasons before drawing. Last month, OpenAI began testing ads inside image generation in the US.

Get the TNW newsletter

Get the most important tech news in your inbox each week.