Static images are no longer the only thing you can make with ChatGPT. OpenAI’s new Images 2.6 update adds GIF support to ChatGPT, Codex, and the API, letting you turn your creations into short animations.
OpenAI says people generate around 4 billion images a week across these platforms, or more than half a billion every day. With that many images being made, adding motion feels like the obvious next step.
Turn a doodle into an animated character
For me, GIF support is the standout feature of this update. Marketers can animate a brand mascot for a campaign, online sellers can make product photos play on repeat, and social media users can create their own reaction GIFs and memes.
Animators can also plan character movement with animated storyboards before shooting any video. You don’t need a finished design to start. A basic doodle is enough for ChatGPT to turn into an animated character for videos or games.
How much faster is Images 2.6?
OpenAI says the new model generates images 50% faster than its predecessor. It also keeps characters, products, and brand elements looking the same across multiple edits. If you have ever watched an AI tool change your character’s face after a small tweak, you know why this matters.
Developers can pick between two new models. GPT-Image-2.6 Flare is built for speed and can create an image in as little as six seconds on its low-quality setting. It also arranges posters and product images more neatly. GPT-Image-2.6 Sunburst focuses on detail and produces sharper images that stay consistent across edits on medium or high quality.
The API also supports image markup, letting developers highlight the exact area they want changed, and SVG export for use on websites, apps, and in print. A new complexity setting controls how much planning the model does before generating, and lower values work well for small edits.
If you already use ChatGPT for social posts or product images, you can now create animated versions in the same place and get your results in half the time.