
Sora can generate up to 1080p resolution videos up to 20 seconds long, and allows users to remix AI-generated content with their own assets.
Just days after Google expanded the availability of Veo, its video-generative artificial intelligence tool for Vertex AI customers, OpenAI made Sora Turbo open to its paid subscribers as a standalone product on Monday.
However, just hours after its release, OpenAI temporarily disabled sign-ups for Sora Turbo due to heavy traffic. The video-generative artificial intelligence tool has been in preview mode since February 2024.
we are launching sora today, and we made a new product to go with it.
— Sam Altman (@sama) December 9, 2024
if you have an openai plus or pro account, you can generate videos. anyone can view them.
it will take some time to roll out, but by the end of the day it should be available at https://t.co/VZBcJFqChS
Sora Turbo has been developed by a team led by Aditya Ramesh, Vice-President, Research, OpenAI, and can be accessed from sora.com. Ramesh said Sora is pushing the boundaries of visual generation, which will require breakthroughs both in ML and HCI.
“One of the most exciting things to me about this product is how easy it is to co-create with others; it feels like an interesting new thing. This is early, think of it like GPT-1 for video, but I already think the feed is so compelling,” said Sam Altman, CEO, OpenAI, on X.
The company said Sora Turbo is significantly faster than the previous model it previewed in February, and that it has also developed a new user interface to make using this tool easy and intuitive
OpenAI Sora Turbo is a cutting-edge text-to-video AI model — users can give it text prompts, and it will generate a video based on the description. Trained on massive text and visual data, Sora can understand the relationship between words, images, and videos, and can generate realistic-looking videos in seconds. It is also one of the first AI-powered video generation models available to the public.
OpenAI Sora Turbo can generate up to 1080 p resolution videos, which are up to 20 seconds long, in both vertical and horizontal formats. Users will also have the option to remix these AI-generated videos with their assets to extend, blend, and remix them as needed.
ChatGPT Plus subscribers, who pay $20 a month, can generate up to 50 videos at 480 p resolution per month or lesser videos in 720 p resolution, and ChatGPT Pro users, who pay $200 a month, will have 10 times more usage at higher resolution and longer duration.
OpenAI has also confirmed it will offer tailored pricing to access Sora Turbo starting early next year.
All the videos generated by Sora Turbo will include Coalition for Content Provenance and Authenticity (C2PA) metadata, and to ensure transparency, these videos will also include a visual watermark. To prevent abuse, OpenAI has blocked child sexual abuse materials and sexual deepfake video generation.