OpenAI Images 2.6: faster generation, GIFs and SVG export

On 4 November 2026, OpenAI introduced Images 2.6 across ChatGPT, Codex and the API — a release aimed less at a single spectacular demo and more at everyday creative tempo. Generation is fifty percent faster than before. In the API, developers can choose GPT-Image-2.6 Flare, which can return images in as little as about six seconds at low quality, or GPT-Image-2.6 Sunburst, tuned for higher quality and stronger consistency across edits at medium and high quality settings.

The feature list reads like a studio wish list: GIF animation; image markup now in the API as well as ChatGPT; SVG export for work that needs to travel into websites, apps or print; and a new complexity parameter that adjusts how much planning the model invests in a request. OpenAI says people now create four billion images a week across ChatGPT, Codex and the API — on average more than half a billion a day. Images 2.6 is available globally on all ChatGPT and Codex subscription tiers.

Early partners are already folding the model into their tools. OpenArt’s Coco Mao called GIFs “the native language of social media,” arguing that GPT-Image-2.6 lets creators speak that language from a single prompt. Town, a startup building an AI work assistant, used the API to explore animated characters for everyday workplace interactions. At Figma Weave, Creative Director Itay Schiff said the model brings speed and creative control together while helping preserve the details that define a character, product or brand — and that Weave is adding it as another model for creative production inside reusable workflows.

For makers who already live between sketchbooks and screens, the practical shift is iteration. Faster rounds, markup that points edits exactly where they belong, SVG paths that leave the chat window ready for production, short GIF loops for campaigns or storyboards. FeelArtsy’s own cover practice is AI-assisted illustration under studio direction; tools like these change the pace of that craft without replacing the editorial eye that decides what belongs on a page.

Images 2.6 will not settle debates about authorship or taste. It does, however, make one claim hard to ignore: image models are becoming infrastructure. When half a billion pictures appear each day, the interesting question is no longer whether people will use them — it is how carefully they will edit, credit and compose them.