OpenAI has introduced Images 2.6 across ChatGPT, Codex and its API, adding GIF creation, more precise image editing and controls for developers. The company says the model generates images 50% faster than its predecessor.
In the API, developers can choose between GPT-Image-2.6 Flare and Sunburst. Flare is designed for speed and can generate an image in as little as six seconds when quality is set to low. It also improves layout quality for posters, product imagery and other professional visuals. Sunburst targets tasks requiring greater detail, with higher-quality generation and improved consistency across edits when quality is set to medium or high.
Images 2.6 can turn images into animated GIFs, including brand mascots, characters and product shots. The company also describes uses such as social media reactions and animated storyboards for exploring character movement. A simple sketch can serve as the starting point for an animated character.
The API now supports image markup, which was already available in ChatGPT, allowing teams to identify specific areas for changes. Developers can also export designs as SVG files and use a new complexity setting to control how much planning the model applies. OpenAI recommends low complexity for quicker edits and medium or high settings for scenes with more elements or constraints; image quality separately controls visual detail and fidelity.
OpenAI says people create four billion images per week across ChatGPT, Codex and the API, averaging more than half a billion per day. Images 2.6 is available globally in ChatGPT and Codex across all subscription tiers on desktop, mobile and web. GPT-Image-2.6 Flare and Sunburst are available through the API.
Sources: OpenAI News.