OpenAI has introduced animated GIF capabilities to ChatGPT through its latest image generation model, Images 2.6. The feature enables users to create GIFs directly from written descriptions or uploaded images, then customize playback speed and modify individual frames before sharing. This represents the first time ChatGPT has offered one-step GIF creation.
The new model also delivers faster image generation performance across ChatGPT. The rollout is reaching all users of ChatGPT and Codex on web, desktop, and mobile platforms.
For developers, OpenAI is making available two distinct API models: GPT-Image-2.6 Flare, optimized for speed, and GPT-Image-2.6 Sunburst, designed for quality output. This dual-model approach mirrors the strategy OpenAI established with Images 2.5 in September, when Adobe integrated one of its models into Firefly.
The GIF feature arrives months after OpenAI discontinued Sora, its AI video generation application, in March.
Usage of OpenAI's image tools continues to accelerate. The company reports that users now generate more than 4 billion images weekly across ChatGPT and its API, a jump from 3 billion in September. These figures encompass all image creation activity, not exclusively GIF production.
OpenAI has maintained a steady cadence of image tool improvements throughout the year. Images 2.0, which launched in April, was the company's first image model capable of reasoning before generating visuals. Last month, OpenAI began piloting advertisements within the image generation interface in the United States.
Source: The Next Web
