OpenAI adds GIF creation to ChatGPT with Images 2.6
The rollout includes frame editing and playback controls, while developers receive separate API models focused on speed and quality.
OpenAI is introducing Images 2.6 and animated GIF creation in ChatGPT, The Verge reported. The feature lets users generate animations from either text prompts or uploaded photos, adding a capability ChatGPT previously could not complete in a single step.
Users can adjust playback speed, inspect individual frames, edit them and share the resulting GIF. OpenAI says the new model also speeds up image generation. The rollout covers all ChatGPT and Codex users across mobile, desktop and web platforms.
For developers, OpenAI is adding GPT-Image-2.6 Flare, which prioritizes speed, and GPT-Image-2.6 Sunburst, which focuses on quality. The two-model approach continues a split introduced with Images 2.5 in September.
OpenAI puts weekly image creation across its API and ChatGPT at more than 4 billion, compared with 3 billion in September. Those company-reported figures include all image generation rather than GIFs alone.
The GIF feature follows the March shutdown of OpenAI’s Sora video app. Other changes to its image tools include the April launch of Images 2.0, described as its first image model to reason before drawing.
