Sora

video

by OpenAI · Updated December 1, 2025

Sora is OpenAI's groundbreaking text-to-video generation model that can create realistic and imaginative video scenes from text descriptions. It understands complex motion, camera work, and can generate videos up to a minute long with consistent characters and settings.

Best For

Cinematic scenesShort filmsProduct videosCreative storytellingVisual concepts

Prompting Tips

1Describe the scene as if writing a film script — include setting, action, and mood
2Specify camera movement: tracking shot, dolly zoom, crane shot, handheld
3Mention lighting conditions and time of day for atmosphere
4Describe motion explicitly: "slowly walks", "wind blowing through hair"
5Keep characters consistent by describing them in detail once
6Shorter, focused prompts tend to produce more coherent results

Syntax & Constraints

Natural language prompts describing the video scene. Specify duration, aspect ratio, and resolution. Camera movements and transitions can be described naturally.

Example Prompts

🏔 Landscape & Scenery

Aerial mountain sunrise

A cinematic aerial shot slowly flying over a misty mountain range at sunrise. Golden light breaks through clouds, illuminating snow-capped peaks. Camera gradually descends toward a pristine alpine lake reflecting the mountains. Smooth, slow motion.

aerialcinematiclandscape

👤 Portrait Photography

Portrait video close-up

Close-up portrait shot of a woman looking directly at camera, soft natural window light illuminating her face. A gentle smile forms as she tilts her head slightly. Shallow depth of field, warm color grading, subtle film grain. 4 seconds.

portraitclose-upnatural-light