ChatGPT can make animated GIFs, with controls for speed and frames
GIF creation arrives in ChatGPT
ChatGPT can produce a GIF from a text request or a picture supplied by the user. Previously, the chatbot did not offer a built-in, one-step way to make these animations.
After generation, users can alter how quickly the animation plays and work on frames individually. They can also share their completed GIF with others.
A prompt demonstrates the feature
OpenAI showed an example in which a user asks for a dancing cat GIF. The demonstration also lets the user adjust playback speed and inspect the animation one frame at a time.
The update makes those editing tools part of the GIF workflow in ChatGPT. The company’s example illustrates how a prompt can produce an animation that users can then refine.
Images 2.6 rollout and API tools
The GIF feature is arriving alongside OpenAI’s Images 2.6 model, which the company says generates images faster. OpenAI says users create over 4 billion images each week across ChatGPT and its API, compared with 3 billion weekly in September.
Images 2.6 is rolling out to ChatGPT and Codex users on the web, desktop, and mobile apps. OpenAI is also introducing GPT-Image-2.6 Flare and Sunburst in its API for faster, higher-quality image generation.
Key points
- ChatGPT can make GIFs from prompts or uploaded photos.
- Users can change animation speed, edit frames, and share GIFs.
- Images 2.6 is rolling out across ChatGPT and Codex platforms.
- OpenAI reports more than 4 billion images generated weekly across ChatGPT and its API.