official announcement

Sora research demonstrates text-conditioned video generation

OpenAI disclosed a diffusion-transformer video model with selected demonstrations and acknowledged simulation limitations.