Subscribers to ChatGPT Pro and Plus can now access OpenAI Sora, the much-anticipated video generation model. Starting later today, users in the US and most other supported countries can experiment with this innovative tool, OpenAI confirmed on Monday. The public release follows an early preview earlier this year. Sora now operates with a more robust model compared to the initial showcase in February. Known as Sora Turbo, this version boasts significant speed improvements.
However, OpenAI notes that the model still faces challenges, such as creating realistic physics and handling intricate actions in extended durations.
Upon visiting Sora’s dedicated landing page, users encounter a video feed showcasing clips generated by others.
Exploring Sora’s Features
Selecting a video reveals the prompt used for its creation. Additionally, users can remix, blend, or re-cut the clips into their projects. For this release, OpenAI has restricted video output to a maximum resolution of 1080p and a duration of 20 seconds.
ChatGPT Plus subscribers can create up to 50 videos monthly at 480p. Alternatively, they can opt for fewer, shorter videos at 720p. Pro subscribers enjoy significantly expanded capabilities, generating 10 times more content at higher resolutions and longer lengths. OpenAI plans to roll out tailored pricing options for various user needs by early next year.
Each video includes a visible watermark and C2PA metadata to ensure content authenticity and traceability. To maintain safety, OpenAI prevents Sora from producing child sexual abuse materials (CSAM) and explicit deepfakes. Furthermore, the platform restricts video uploads featuring people until it strengthens safeguards against misuse.
This launch positions OpenAI Sora release as a pioneering solution in video generation technology, with exciting potential for creativity and practical applications. The OpenAI Sora video generation model’s debut promises to reshape how users approach video creation in digital spaces.