OpenAI announced on November 4, 2026 the launch of Images 2.6, available globally in ChatGPT, Codex and the API. The update makes image generation 50% faster and adds support for animated GIFs, reusable reference images and greater consistency across successive edits.
Key new features
Images 2.6 lets users create GIFs from prompts or existing images, useful for brand mascots, social media reactions or character storyboards. It also improves preservation of character, product and brand details during multiple editing rounds.
In the API, OpenAI offers two models: GPT-Image-2.6 Flare, optimized for speed (generating images in as little as six seconds at low quality), and GPT-Image-2.6 Sunburst, aimed at higher quality and consistency in medium or high quality edits. Both are available at pricing aligned with the previous generation.
New developer features include SVG export, area markup for precise editing and a complexity parameter that controls how much planning the model does before rendering.
Availability and context
The feature is already available on all ChatGPT and Codex subscription tiers, on desktop, mobile and web. OpenAI says users generate about 4 billion images per week on the platform.
Partners such as OpenArt, Figma Weave and Picsart highlighted gains in speed and consistency for creative and production workflows.
The company published a system card with safety evaluations. More details are in the official announcement.
This article was produced with the assistance of artificial intelligence and reviewed for editorial accuracy.
Por Grok
