OpenAI moved its Sora video generation model out of preview mode to a commercial standalone product, part of plan to better compete against AI rivals Anthropic, Google and Elon Musk’s xAI.

The company explained Sora can create realistic videos featuring multiple characters and detailed backgrounds from text.

It stated the AI model “understands and simulates reality”, which it argued is an “important step towards developing models that can interact with the physical world”.

ChatGPT developer OpenAI stated the latest version, Sora Turbo, is significantly faster than the model it previewed in February.

It is available to existing ChatGPT Plus and Pro customers, except in the UK, Switzerland and the European Economic Area.

OpenAI stated it is working to expand availability in the coming months.

Sora can generate 20-second videos with up to 1080p resolution in widescreen, vertical or square aspect ratios. Users can bring their own assets “to extend, remix and blend, or generate entirely new content from text”, OpenAI stated in a blog.

“We’ve developed new interfaces to make it easier to prompt Sora with text, images and videos. Our storyboard tool lets users precisely specify inputs for each frame”.

The company cautioned the current version of Sora comes with “many limitations”.

“It often generates unrealistic physics and struggles with complex actions over long durations,” OpenAI stated, adding it is still “working to make the technology affordable for everyone”.

The text-to-video model builds on past research OpenAI conducted for its AI image generator DALL-E and its open API product.

In October, OpenAI established a $4 billion revolving line of credit with a consortium of banks and announced a $6.6 billion fundraising round with prominent technology companies.

Last month, it announced a ChatGPT Search model.