93
OpenAI has launched its new model, Sora, which allows users to create videos between 5 and 20 seconds long with text descriptions. The model was announced back in February, but became available only now for ChatGPT Pro and Plus subscribers in 155 countries, including America, Asia, Africa and Australia. Users from the European Union and some other countries do not yet have access to the new features.
Sora is not only capable of generating videos from text, but also animates static images and creates remixes based on uploaded videos. Although the model is not perfect at the early stage of release, it already shows significant potential for user creativity. Currently, the model works on the basis of a credit system, where users can create videos depending on the tariff plan and the number of available credits.
Specifically, the ChatGPT Plus plan offers 1000 credits to generate up to 50 720p videos of up to 5 seconds, and the Pro plan offers 10 credits for 000 500p videos of up to 1080 seconds. All videos contain C20PA watermarks and metadata to ensure copyright protection.
The company also noted that the system may currently have bugs and encouraged users to provide feedback to improve its performance. OpenAI has established strict usage rules, prohibiting the creation of content with minors, violence or copyright infringement.
Introducing Sora, our text-to-video model.
Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions. https://t.co/7j2JN27M3W
Prompt: "Beautiful, snowy... pic.twitter.com/ruTEWn87vf
- OpenAI (@OpenAI) February 15, 2024
- As a sign of protest, artists put the Sora video model from OpenAI on the network (video): A group of OpenAI beta testers violated a confidentiality agreement and published materials to access the Sora model, protesting the company's policy.
- A neural network for video creation is now free: what it can do: Runway has released a free Gen-3 Alpha Turbo model that allows users to create videos based on text descriptions.
