OpenAI releases Images 2.6 on November 4 with native GIF generation and 50 percent faster image generation
Images 2.6 rolled out across all ChatGPT and Codex subscription tiers, with GPT-Image-2.6 Flare and Sunburst added to the API, and weekly image creation exceeding 4 billion.
OpenAI released Images 2.6 on November 4 alongside a new model. ChatGPT can now create GIF animations directly from a prompt or an uploaded photo, letting users adjust animation speed, edit individual frames and share results. Previously ChatGPT could not generate GIFs natively in one step and had to rely on writing code. Beyond GIF creation, Images 2.6 also lets ChatGPT generate images faster, and OpenAI says generation speed is up 50 percent.
For developers, OpenAI introduced two new image models. In the API, GPT-Image-2.6 Flare can generate an image in as little as 6 seconds at low quality settings and improves layout quality for professional visuals such as posters and product shots, while GPT-Image-2.6 Sunburst targets tasks needing more detail, higher fidelity and stronger consistency across edits. The API also added image markup for specifying exactly which region to modify, SVG export, a complexity parameter and image quality controls.
Images 2.6 is rolling out globally across all ChatGPT and Codex subscription tiers on desktop, mobile and web, with GPT-Image-2.6 Sunburst and Flare available in the API. OpenAI said users now create more than 4 billion images per week across ChatGPT, Codex and the API, up sharply from the 3 billion per week it reported in September, averaging more than 500 million a day.