OpenAI has released Sora Turbo, a multimodal AI-powered video generation model that can create realistic videos from text.

Currently available for ChatGPT Plus and Pro users, Sora Turbo is an enhanced and significantly faster version of Sora, which the firm launched in February.

“Sora serves as a foundation for AI that understands and simulates reality, an important step towards developing models that can interact with the physical world,” the firm said in a statement.

The launch includes a range of new functions, including the ability to generate videos that last up to 20 seconds with a maximum 1080p resolution across widescreen, portrait or square formats. Users can also bring in their own resources to extend, remix and merge, or generate completely new content from text.

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

Sora is included in ChatGPT's Plus account at no extra cost, with a basic feature that allows users to generate up to 50 videos per month at 480p resolution or less at 720p.

A Pro plan includes 10x more usage, higher resolutions, and longer durations.

The firm said it’s currently working on tailored pricing for different types of users, which it plans to release in 2025.

OpenAI warned that the new version still has many limitations.

"It often generates unrealistic physics and struggles to handle complex actions of long duration,” said the company.

It added that not all content will be available and that the platform will block forms of abuse, such as child sexual abuse and sexual deepfakes.

“Uploads of people will be limited at launch, but we intend to roll the feature out to more users as we refine our deepfake mitigations,” it added.

---