OpenAI's Images 2.6 adds GIF creation to ChatGPT and Codex
The new model turns prompts or uploaded photos into editable GIFs and generates images about 50 percent faster, OpenAI says.
OpenAI released Images 2.6 for ChatGPT, Codex and the API, adding the ability to generate GIFs from a text prompt or an uploaded photo.
Users can edit a generated GIF before saving or downloading it, including changing the loop speed and adjusting individual frames. Uploaded photos or sketches can also be saved and tagged as reference images for reuse later.
According to OpenAI, Images 2.6 generates images about 50 percent faster than the previous version. The model is available to all ChatGPT and Codex users, though image limits depend on the subscription tier.
Two models for API developers
Developers working through the API can choose between Flare and Sunburst, depending on whether they prioritise speed, detail or consistency.
OpenAI has previously said people generate more than 500 million images each day, on average, according to engadget.com.
Quick answers
What is new in OpenAI's Images 2.6?
Images 2.6 can create GIFs from a prompt or an uploaded photo, and those GIFs can be edited before saving or downloading.
What is the difference between Flare and Sunburst in Images 2.6?
API developers can pick between the two models based on their speed, detail and consistency needs.
Can I use Images 2.6 on the free ChatGPT tier?
Images 2.6 is available to all ChatGPT and Codex users, but image limits depend on the subscription.