Sora generates video from text or image prompts. The Feb 2024 demo was a watershed moment for AI video — quality jumped roughly two years ahead of competitors overnight. Currently bundled in ChatGPT Pro ($200/mo). Strengths: best raw quality for single-shot work, longest clip lengths (60 seconds), strong camera-control vocabulary. Weaknesses: standalone generator without an editing suite (Runway wins on workflow), character consistency across clips still imperfect, no synced dialog. By 2026, used heavily for short-form social, B-roll, and concept-stage filmmaking.
מילון
מה זה Sora?
OpenAI's text-to-video model, launched publicly in early 2024. Industry-leading raw video quality, 60-second clips on the Pro tier.
מונחים קשורים
Video Generation
AI that creates video clips from text prompts, reference images, or existing video — Sora, Runway, Veo, Kling, and Pika lead the category.
Diffusion Model
The AI architecture behind most image generators (Stable Diffusion, DALL-E, Midjourney) — generates images by progressively denoising random noise.
OpenAI API
Programmatic access to OpenAI's models (GPT-5, o-series, embeddings, DALL-E, Whisper, Sora). The most widely-used commercial LLM API.
חזרה ל- מילון ה-AI