Sora is OpenAI's text-to-video generation model capable of producing high-definition, physically plausible video clips up to one minute long from text prompts. It understands complex scenes, camera movements, and temporal coherence. Sora represents a major leap in generative video and is aimed at filmmakers, educators, and creative professionals.