Technology

Sora

OpenAI's Sora is a diffusion transformer model that generates hyperreal, up to 20-second video clips from natural language text prompts.

Sora, developed by OpenAI, is the premier text-to-video generative AI model: it converts natural language prompts into high-fidelity, dynamic video clips. The current Sora 2 model leverages a diffusion transformer architecture to generate content up to 20 seconds long at 1080p resolution (Pro tier), demonstrating a strong grasp of 3D space, physics, and long-range scene continuity. It handles complex scenes, such as 'a stylish woman walking down a neon-lit Tokyo street,' and can also animate static images or remix existing video. Access is currently available to users with a ChatGPT Plus or Pro subscription.