OpenAI Rolls Out Images 2.6 Model, Enabling Animated GIF Generation in ChatGPT

OpenAI has upgraded ChatGPT with its latest Images 2.6 model, bringing the ability to generate animated GIFs directly from text prompts and uploaded photos. The update is rolling out to all ChatGPT and Codex users across web, desktop, and mobile applications.

Flipbook Animations and Frame-Level Control

With the Images 2.6 model, users can request animated GIFs by typing a text prompt or submitting an existing image. The tool allows users to fine-tune the output by adjusting the animation speed or editing individual frames.

To demonstrate the capability, OpenAI showcased a prompt asking: "Can you make a gif of a cat dancing?" The model produces slight character variations that are strung together flipbook-style to form a smooth GIF animation. A key improvement in Images 2.6 is its image consistency across frames, keeping characters recognizable throughout the animation—a task that earlier models struggled with due to excessive frame-to-frame variations.

Faster Generation and Developer APIs

Beyond adding GIF creation, the Images 2.6 update is designed to increase overall image generation speed within ChatGPT.

For developers, OpenAI is releasing two updated API model variants based on the new architecture:

• GPT-Image-2.6 "Flare": Optimized for efficient, fast image generation.
• GPT-Image-2.6 "Sunburst": Designed for higher-quality image outputs.

Both model names trace back to variants first introduced alongside Image 2.5.

OpenAI reported that image generation activity across ChatGPT and its API has reached new highs. Users now generate well over four billion images per week, representing an increase of one billion weekly images compared to figures from September.