by OpenAI
OpenAI's video generation model producing cinematic-quality clips from text prompts.
Sora by OpenAI is a video generation model known for strong cinematography literacy, physics-aware scene generation, and natural motion. Sora 2 (released late 2025) added synchronized dialogue/sound effects, higher resolution output, and improved temporal consistency. Available through ChatGPT Plus/Pro and the API.
Sora responds best to structured, visual descriptions:
Medium shot of an elderly watchmaker at his workbench, carefully
adjusting tiny gears with tweezers. Warm tungsten lighting from
a desk lamp creates dramatic shadows. His hands are steady and
precise. Camera slowly pushes in to an extreme close-up of the
watch mechanism. Shallow depth of field.
Short, vague prompts give Sora too much room to improvise. Describe exactly what the shot should achieve — as if briefing a cinematographer.
For Sora 2+, keep visual description and spoken dialogue in separate blocks for best results.
| Parameter | Description |
|---|---|
| duration | Video duration in seconds (up to 10s) |
| resolution | 480p, 720p, 1080p |
| aspect_ratio | 16:9, 9:16, 1:1 |
| style | cinematic, natural, animated |
| seed | Reproducibility seed |
Quick tips from the community about what works with Sora right now.
Sign in to share a tip.
No tips yet. Add a tip for this model.