Sora
AI Model
Video
Current
OpenAI's video generator — and, with Sora 2, a social app.
You're looking at an AI model record: who built it, what it can do, and the hardware and data behind it.
Sora (previewed February 2024, released December 2024) generates minute-scale video from text via a diffusion transformer over spacetime patches. Sora 2 (September 2025) added synchronized audio, physical consistency and a TikTok-style app with consent-based 'cameos' — pushing video generation into consumer culture and copyright controversy simultaneously.
How this connects to the rest of the stack
This record's links at a glance — click a group to jump to it.
Developed by
Competes with
Related
Sources
Record reviewed as of 2026-01-15 — see Methodology.