Sora is a text-to-video AI model that's part of a 12-day "ship-mas" product release series. It's now available on Sora.com for ChatGPT subscribers in the U.S. and most other countries, and a new model, Sora Turbo, adds features like creating videos from text, animating images, and combining videos.

OpenAI first revealed its text-to-video AI model, Sora, in February of this year. During the livestream, OpenAI showcased Sora's new discovery with a feed of AI-generated videos from other members of the community.

OpenAI also revealed a “storyboards” feature that allows for video creation based on a series of prompts, as well as turning photos into videos. The tool also lets users tweak Sora’s output with text prompts, along with a way to blend two scenes together using AI.

OpenAI says that videos created with Sora will have watermarks and C2PA metadata indicating they were created with AI. Before uploading images or videos to the app, users must agree to an agreement that the uploaded content does not contain content for people under 18. Failure to do so may result in a ban or account suspension.

For ChatGPT Plus subscribers, OpenAI says it can create up to 50 720p, 5-second priority videos. Users on the $200/month ChatGPT Pro plan can create up to 500 1080p, 20-second priority videos. Additionally, the Pro plan allows subscribers to download videos without watermarks.

Users who don't sign up for ChatGPT can still browse AI-generated video feeds created by other people using Sora. The model is currently available in the US and many other countries, but it will take time to roll out to European countries and the UK.