OpenAI’s Images 2.6 Brings Animated GIF Creation Directly into ChatGPT

OpenAI has rolled out a new image model, Images 2.6, that turns ChatGPT into a one-stop shop for animated content. Users can now generate and edit animated GIFs directly inside the chatbot using nothing more than a written prompt or an uploaded photo, eliminating the need for external animation tools. The update, confirmed by OpenAI and reported across major tech outlets, marks the first time ChatGPT can natively produce GIFs in a single step.

  • GIF creation inside ChatGPT: Users can generate animated GIFs from text prompts or uploaded images, then adjust playback speed, edit individual frames, and share the result without leaving the chat.
  • Faster image generation: Images 2.6 generates images 50% faster than its predecessor, according to OpenAI, and is rolling out to all ChatGPT and Codex users across web, desktop, and mobile.
  • API split: Developers get two variants — GPT-Image-2.6 Flare for speed and GPT-Image-2.6 Sunburst for higher quality and consistency across edits.

From Still Images to Motion

The GIF feature is the headline addition in Images 2.6. OpenAI’s own examples show ChatGPT responding to a prompt like “can you make a gif of a cat dancing?” with a finished animation, alongside controls to change playback speed and inspect the sequence frame by frame. The company positions the tool for marketers, creators, and businesses: animating a brand mascot, producing reaction memes for social media, or building animated storyboards before committing to full video production.

Beyond GIFs, Images 2.6 brings a suite of editing improvements. Image markup, previously available only in ChatGPT, now reaches the API so teams can pinpoint exactly where changes should be made. SVG exports make designs ready for websites, apps, and print, while a new complexity parameter lets developers control how much planning the model does before generating a scene. Lower complexity settings suit quick edits, while higher settings help with scenes containing many elements or constraints.

Two Models, One API

In the API, OpenAI splits Images 2.6 into two distinct models. GPT-Image-2.6 Flare prioritizes speed: when quality is set to low, it can generate images in as little as six seconds, and it also improves layout quality for posters, product imagery, and other professional visuals. GPT-Image-2.6 Sunburst targets detail and precision, delivering higher-quality output and greater consistency across edits when quality is set to medium or high.

The company says the update also accelerates image generation in ChatGPT generally, not just GIFs. The rollout covers all ChatGPT and Codex users on web, desktop, and mobile apps.

Scale and Safeguards

OpenAI reports that users now create more than 4 billion images per week across ChatGPT, Codex, and the API — an average of over half a billion images per day, and a jump from the 3 billion per week reported in September. That volume underscores how central image generation has become to the platform’s usage.

The GIF capability also introduces new safety considerations. OpenAI’s system card notes that GIF generation carries risks not apparent in still images, including sexualized movement and harmful meaning conveyed across frames. The company applies GIF-specific safeguards designed to detect and block policy-violating content, with rules that block some categories more broadly than for still images to account for inferred motion. OpenAI acknowledges that motion-related violations remain challenging to detect and that safeguards may not catch every case.

On the preparedness front, OpenAI says neither GPT-Image-2.6 Sunburst nor Flare crosses the Bio High or Cyber High thresholds, though it applies precautionary mitigations appropriate for a model treated as high biological capability, consistent with its approach for Images 2.0 and 2.5.

The addition of GIFs is less about the format itself and more about OpenAI’s broader strategy of consolidating creative tools inside ChatGPT. By folding animation, editing, and faster generation into a single product, the company reduces the reasons users might leave for third-party apps. For now, the GIF feature arrives at no extra cost — but as OpenAI continues to expand ChatGPT’s capabilities, including experiments with advertising inside image generation in the United States, the implicit trade-off between convenience and dependence on a single platform’s terms and business model becomes harder to ignore.

Sources & References