ChatGPT GIF Maker Arrives With Images 2.6 and Frame Editing
ChatGPT can now turn a text prompt or a photo into an animated GIF with OpenAI's new Images 2.6 model, complete with speed controls and frame editing. Developers also get two new API models.
Hakkı ALKANPublished 3 min read
OpenAI has turned ChatGPT into a GIF maker. With the new Images 2.6 model, a ChatGPT GIF can now start from a simple text prompt or an uploaded photo, and the result comes back as an animated loop instead of a still image. The company announced the update on November 4, 2026, as The Verge first reported.
Until now, ChatGPT could not produce a GIF in a single step. Now, however, a request as casual as OpenAI's own example, "can you make a gif of a cat dancing?", is enough to get a finished animation.
How the ChatGPT GIF Feature Works
Once a GIF is generated, ChatGPT shows a few controls. First, you can change the playback speed. In addition, you can open the individual frames and edit them one by one, which helps when a single frame breaks the motion. Finally, the finished file can be shared directly from the chat.
According to The Decoder, the update also focuses on keeping characters and objects consistent from frame to frame, which has long been a weak spot for AI animation. Meanwhile, OpenAI says Images 2.6 generates regular still images faster as well.
The rollout covers all ChatGPT and Codex users on the web, desktop and mobile apps. In other words, the feature is not limited to paid plans.
Flare and Sunburst Come to the API
For developers, OpenAI is adding two new models, GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst. Flare is tuned for speed, while Sunburst targets higher quality output. As The Next Web notes, this follows the same two-tier split OpenAI introduced with Images 2.5 in September. However, as of Wednesday evening, OpenAI's public API pricing page still listed only the 2.5 versions, so prices for the new models have not been published yet.
The numbers explain why OpenAI keeps investing here. ChatGPT and the API now generate more than 4 billion images per week, up from 3 billion in September. For context, OpenAI recently said ChatGPT search volume has reached about a fifth of Google's, so image tools are part of a much broader push to keep users inside the app.
A Lighter Take on AI Video
GIFs are also a smaller, safer step toward motion than full video. After all, OpenAI shut down the standalone Sora app in March, after the app had faced a deepfake backlash at launch. Short loops built for chats and social media, by contrast, fit naturally into ChatGPT.
The Codex angle matters too. OpenAI has been folding more of its tools together, and Codex now shares usage limits with ChatGPT. As a result, developers can test GIF generation in the same place where they build, before the Flare and Sunburst models reach their own apps.