OpenAI released Images 2.6 on November 4, 2026, an update to the image system inside ChatGPT that adds the ability to create animated GIFs. According to The Decoder, which cited The Verge, users can start from a written prompt or an uploaded photo and receive a short looping animation. (The Decoder (opens in a new tab); Unite.AI (opens in a new tab))

The company said the update is rolling out globally to ChatGPT and Codex users across all subscription tiers, on the web, desktop and mobile apps, TestingCatalog and ETV Bharat reported. (TestingCatalog (opens in a new tab); ETV Bharat (opens in a new tab))

How the GIF feature works

The model builds a GIF much like a flipbook. It generates a series of closely related still images and plays them in sequence. In OpenAI's own example, a short request for a GIF of a dancing cat produced several slightly different versions of the same character strung together, The Decoder reported. (The Decoder (opens in a new tab))

After a GIF is generated, users can adjust the playback speed, inspect and revise individual frames, and then share the result. ETV Bharat reported that a rough sketch can also serve as the starting point. In its blog post, OpenAI wrote that "From personal reactions to brand mascots and product visuals, GIFs give people and businesses a playful way to communicate." (ETV Bharat (opens in a new tab); Unite.AI (opens in a new tab))

The feature depends on consistency. The Decoder noted that earlier OpenAI models would likely have changed a character's look too much from frame to frame for an animation to hold together, which makes GIFs a visible test of how far the company's image models have come in a short time.

Saved references and consistent edits

Images 2.6 also lets users keep reference images of themselves, their pets or recurring characters and reuse them later without uploading them again. OpenAI summarized the idea as "Save once. Create again and again," according to TestingCatalog. The company said the update is meant to keep characters, products and brand details consistent across repeated rounds of editing. (TestingCatalog (opens in a new tab); ETV Bharat (opens in a new tab))

Flare and Sunburst for developers

For developers, OpenAI introduced two API models, GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst. Flare is aimed at speed and layouts such as posters and product images, and the company said it can produce a low-quality image in as little as six seconds. Sunburst is designed for more detail and for keeping edits consistent at medium or high quality settings. (TestingCatalog (opens in a new tab); ETV Bharat (opens in a new tab))

The API also gains the image markup tool already available in ChatGPT, which lets teams point to the exact area of a picture they want changed, along with SVG export for websites, apps and print. A new complexity setting controls how much planning the model does; OpenAI recommends low for most quick edits and medium or high for crowded scenes, TestingCatalog reported. OpenAI said the release generates images 50 percent faster. (TestingCatalog (opens in a new tab); ETV Bharat (opens in a new tab))

OpenAI also said people now create well over 4 billion images a week across ChatGPT and its API, about one billion more than in September. The previous version, Images 2.5, arrived on September 8, 2026, according to Unite.AI. (The Decoder (opens in a new tab); Unite.AI (opens in a new tab))

In NeuraFeed's reading, GIFs move ChatGPT further into everyday messaging, where reaction images and short loops are shared constantly. Pairing a consumer feature with faster tiers for developers suggests OpenAI wants the same model family to serve both casual users and businesses that produce images at volume.