Text-to-video

OpenAI's flagship generative model: transforms text prompts into hyperreal, high-fidelity video clips with synchronized sound.

This technology is a state-of-the-art video model, capable of creating richly detailed, dynamic clips up to 20 seconds long from natural language prompts. It leverages a deep understanding of 3D space and scene continuity to produce unprecedented realism: think 'GPT-3.5 moment for video.' The platform offers two variants: Sora 2 for rapid iteration and flexibility, and Sora 2 Pro for production-quality output, including high-resolution cinematic footage. Key features include synchronized dialogue, sound effects, and a 'Characters' function for casting specific individuals into scenes.

https://openai.com/sora