OpenAI now lets ChatGPT users make GIFs from scratch. The feature arrived Nov. 4, 2026, as part of Images 2.6, an update rolled out for ChatGPT, Codex and the company's API. Users can generate a GIF from a text prompt or a photo, and they can edit it before saving or downloading.
The timing is a playful one. Anyone who has spent years searching for the perfect reaction, from Elmo amid flames to Michael Scott from The Office screaming "Noooooo!" now has a way to skip the clip hunt. OpenAI says people generate more than 500 million images a day on average. The company did not say how many of those are GIFs.
How do you make a GIF in ChatGPT?
Making one is simple. Log in to ChatGPT and type a prompt into the box at the bottom of the screen. OpenAI suggests being specific. Its example is a GIF of a tortoiseshell cat yawning. According to OpenAI's blog post, Images 2.6 should generate images about 50 percent faster than the previous version.
How do you edit a GIF in ChatGPT?
Once the GIF appears, there are several ways to adjust it. The speed control sits in the bottom left corner of the image, where a number such as "1x" shows the current setting. Clicking it changes how fast the loop plays. To make changes beyond speed, click or tap the image and select Edit. A box labeled "Describe edits" will appear, and users can tell ChatGPT what to change. A button labeled "Sheet" shows each individual frame, where users can add comments and markups. When the GIF looks right, the Download and Share buttons handle the rest.
Can you turn a photo into a GIF with Images 2.6?
GIFs are not the only addition in Images 2.6. Users can upload a photo and turn it into a GIF. OpenAI's blog post also says a minimal sketch can serve as a starting point. The company's example is a rough drawing of an owl. Users can also save an image as a reference and tag it for reuse later. According to OpenAI, that works for people, pets and characters, which could help keep a recurring character looking consistent across multiple GIFs.
Developers using the API get a different set of choices. They can pick between two options, Flare and Sunburst, depending on how fast, detailed and consistent they need their image generation to be. OpenAI did not spell out the tradeoffs between the two in the material provided.
Images 2.6 is available to everyone through ChatGPT and Codex, and no subscription is required to use it. A user's subscription level still matters, though. It determines how many images a person can generate.
Dieser Artikel wurde mit Unterstützung von KI verfasst.
News Factory APP - agentische News für besseres SEO & AEO.