OpenAI is adding animated GIF creation to ChatGPT, powered by a new image model called Images 2.6. Users can generate a GIF from a text prompt or from an uploaded photo, then refine the result — The Verge reports the update allows adjusting the GIF's animation, while The Next Web notes users can edit individual frames.

The feature ships alongside developer access: OpenAI is releasing two new API models with the Images 2.6 rollout, extending the animation capability beyond the chatbot interface and into third-party apps.

Strategically, this fills an obvious gap. Static image generation has been table stakes for AI chatbots for years, and full video generation sits at the opposite end of the cost and complexity spectrum. Animated GIFs occupy the middle — lightweight, shareable, and native to how people actually communicate in chats and social feeds. Giving ChatGPT a one-prompt GIF button targets everyday usage rather than professional creative work.

The frame-editing angle is the more interesting technical wrinkle. Treating a GIF as an editable sequence rather than a single opaque output suggests Images 2.6 maintains some consistency across frames — the hard problem in any animation model. How well that holds up outside OpenAI's demos will determine whether this is a toy or a genuinely useful tool.