ChatGPT is gaining animated GIF creation from text prompts or uploaded photographs, with controls for playback speed and individual frames. OpenAI is rolling the feature out with Images 2.6 and has also released developer models supporting animated GIF generation. The company says its broader image tools now produce more than four billion images a week.
An animated image made in ChatGPT can now be adjusted without leaving the conversation. Users can change how quickly a GIF plays, inspect its separate frames and edit individual frames before sharing the result. OpenAI is adding these controls with Images 2.6. A written description or an uploaded photograph can provide the starting point for the animation.[1]
OpenAI has also released GPT-Image-2.6 Flare and Sunburst for developers, with animated GIF generation among the supported capabilities. The rollout of Images 2.6 covers ChatGPT and Codex on the web and in desktop and mobile applications. Previously, ChatGPT could not produce a GIF natively in one step. The new workflow adds moving images to the generation and editing tools.[1], [2]
OpenAI says the update also speeds up image generation. The company puts weekly production across ChatGPT and its application programming interface at more than 4 billion images, up from the 3 billion weekly figure it announced in September. These are the company’s usage figures. Its demonstration of the new animation controls uses a dancing cat, with speed adjustment and a frame-by-frame view available alongside the generated result.[1]
