Video generation is more complex than text or image generation, and in fact the new model, although advanced, is not perfect and could be inaccurate. Sora can be activated by all users who have a Plus or Pro subscription.
It is the second announcement from OpenAI, after the subscription to ChatGPT pro for $200. By the end of the year, the leading artificial intelligence company has promised to launch 12 new products and has now presented Sora. “Our model can create realistic videos starting from the text. Sora is a fundamental point for the development of artificial intelligence, capable of understand and simulate realityan important step towards creating models capable of interact with the physical world“OpenAi explained in a note on the site.
“We have developed a new version of Sora, Sora Turbo, which is significantly faster than the model previewed in February. We’re releasing it today as a standalone product on Sora.com for ChatGPT Plus and Pro users, at no additional cost. European usersHowever, they will have to wait. OpenAI will be able to distribute the new model when it is in compliance with community regulations on privacy andartificial intelligence.
How Sora Turbo works
Sora is a stable diffusion model that generates a video starting from a text. “The AI can generate complex scenes with multiple characters, specific types of movement, and accurate subject and background details. The model has a deep understanding of language, which allows it to accurately interpret instructions.”
How to delete your name from ChatGPT forever
Text-to-video “multimodal models” actually exist already, Sora promises to generate more precise and longer videos. “Users can generate videos up to 1080p resolution, up to 20 seconds long, and in widescreen, vertical or square. You can also extend, remix and merge sequences, or generate completely new content,” explains the company.
The new Sora Turbo model also has new interfaces to make it easy to send text, images and videos and the storyboard tool that allows users to accurately describe the inputs for each frame. “Sora is included as part of your Plus account at no additional cost. You can generate up to 50 videos at 480p resolution or less 720p video every month.”
The limits of the new OpenAI model
Video generation is more complex than text or image generation, and in fact the new model, however advanced it’s not perfect and may be inaccurate. As OpenAI explains: “The version of Sora we are distributing has many limitations. It often generates images that they are not realistic and has difficulty creating complex, long-lasting sequences. Although Sora Turbo is much faster of the February preview, we are still working on making the technology accessible to all.”
Second Reece Haydensenior analyst at market research firm ABI Research, these AI models could have a great impact on the marketsdigital entertainment creating new personalized content streamed. “We are introducing our video generation technology now to give the company time to explore the possibilities and develop together safety rules and measures that ensure responsible use as the sector progresses.”
