The short answer
OpenAI’s original Sora showcase paired videos with prompts for scenes including woolly mammoths crossing a snowy meadow, a cinematic view of Big Sur, a space-traveler trailer, and a papercraft coral reef. The examples show how prompts can describe a subject, action, setting, camera view, lighting, or style. They’re creative references, not guarantees of repeatable results.
Production caveat: OpenAI says the Sora web and app experiences were discontinued on April 26, 2026, and the Sora API was discontinued on September 24, 2026. As of November 12, 2026, these examples are references to past Sora outputs—not evidence that Sora is available for a new API workflow. (OpenAI Help Center)
OpenAI Sora video examples and prompts
The prompts and clip links below are listed in a third-party Sora video collection on GitHub, which groups them under a heading attributing them to OpenAI’s official website. That label is the collection’s attribution; it is not independent verification of each clip’s provenance or prompt-to-video match. The direct links are identifiable viewing targets, but current playback has not been confirmed.
- Woolly mammoths in a snowy meadow — Prompt excerpt: “Several giant wooly mammoths approach treading through a snowy meadow…” View the listed clip.
- Big Sur coastline — Prompt excerpt: “Drone view of waves crashing against the rugged cliffs along Big Sur’s garay point beach.” View the listed clip.
- Space-traveler movie trailer — Prompt excerpt: “A movie trailer featuring the adventures of the 30 year old space man wearing a red wool knitted motorcycle helmet…” View the listed clip.
- Fluffy monster and melting candle — Prompt excerpt: “Animated scene features a close-up of a short fluffy monster kneeling beside a melting red candle.” View the listed clip.
- Papercraft coral reef — Prompt: “A gorgeously rendered papercraft world of a coral reef, rife with colorful fish and sea creatures.” View the listed clip.
These are examples from a collection, not a benchmark of typical output quality. A listed clip does not establish that a prompt will reproduce the same result or that the same model remains available. OpenAI’s original Sora showcase says its showcase videos were generated directly by Sora without modification; that general statement does not independently verify the third-party collection’s individual links or attributions.
What you can learn from the prompts
The examples show different kinds of detail a prompt may include. The mammoth excerpt names a subject and action; its full listed prompt adds the snowy landscape, distant mountains, afternoon light, and a low camera view. The Big Sur prompt describes moving waves, a lighthouse, and sunset light. The monster prompt adds visual treatment and mood.
These are observations about example prompts, not required syntax. OpenAI’s Sora 2 prompting guide discussed the trade-off between giving more direction and leaving room for variation, but the guide is marked archived and may refer to outdated models or APIs. With Sora discontinued, don’t treat it as current operating guidance. (OpenAI Cookbook)
When studying or adapting a prompt, consider:
- Subject and action: Who or what is in the shot, and what happens?
- Setting: Where does it take place, and what surrounds the subject?
- Camera and composition: Is the view close, wide, low, or aerial?
- Light and mood: What time of day, colors, or emotional tone are described?
- Style: Does the prompt suggest a cinematic, animated, photorealistic, or material-specific look?
These are creative questions, not a promise that a model will follow every detail.
Can you use these examples with the Sora production API?
Not as a new Sora API workflow today. OpenAI’s discontinuation notice says its Sora API was discontinued on September 24, 2026; the Sora web and app experiences ended earlier, on April 26, 2026. The examples can serve as creative references, but they don’t establish current API access, model availability, or a supported production path. (OpenAI Help Center)
The GitHub collection provides prompts and clip links, not API instructions, model identifiers, or current availability details. Treat it as an inspiration gallery rather than production documentation.
Takeaway
Use the examples to study how prompts describe scenes, actions, framing, lighting, and style. The linked clips are listed viewing targets, but their current playback is unconfirmed here, and neither a link nor a prompt proves reproducibility. For production planning, the key point is that OpenAI discontinued the Sora app and web experience in April 2026 and the API in September 2026.
Sources
Awesome-OpenAI-Sora-Videos | GitHub
Woolly mammoths in a snowy meadow — listed clip
Big Sur coastline — listed clip
Space-traveler movie trailer — listed clip
Fluffy monster and melting candle — listed clip
Papercraft coral reef — listed clip
Sora: Creating video from text | OpenAI
What to know about the Sora discontinuation | OpenAI Help Center