OpenAI has introduced GIF creation capabilities to ChatGPT, allowing users to generate looping animations from text prompts or uploaded photos. The feature is part of the new Images 2.6 model update.
Users can adjust the playback speed and edit individual frames of the generated GIFs. This allows for specific corrections within an animation without needing to regenerate the entire file.
Technical specifications for the GIF output include a range of 2 to 16 frames. The default setting consists of 9 frames, with each frame displayed for 200 milliseconds.
The underlying code supports spritesheet-based editing and maintains a still preview during the generation process. The feature is currently accessible via the Images API, though it remains behind a disabled flag.
This release follows the September 8, 2026, launch of the Images 2.5 model, which increased image generation speed by 50 percent. Future updates for the 2.6 model may include 4K resolution support.
