Keep this page open while you write. It has the exact vocabulary Sora 2 understands — shot sizes, camera moves, lighting terms — plus the app settings you need to stop guessing. Every table is copy-paste ready. For full walkthroughs, see 35 best Sora prompts and how to prompt Sora for realistic video.
The shot formula
Sora responds best when you write like a director briefing a crew, not a novelist. Every shot breaks down into five beats, in this order:
| Beat | What to write |
|---|---|
| Subject | Who or what is on screen — be specific about appearance and wardrobe. |
| Action | One clear motion. Not three actions stitched together. |
| Setting | Place + time of day. |
| Camera | Shot size + ONE movement. |
| Light/Style | Mood, lighting term, optional film reference. |
Stack those five beats into one paragraph and you have a complete shot. Grab terms from the tables below, drop them into the formula, and set duration/resolution/aspect ratio in the app before you generate. See Sora prompt templates for fill-in-the-blank skeletons built on this exact formula.
Shot sizes
Shot size controls how much of the subject and setting is visible. Pick one per shot.
| Term | What it frames | Use for |
|---|---|---|
| Extreme wide shot | Subject tiny against a huge environment | Establishing scale, landscapes, opening shots |
| Wide shot | Full subject plus surrounding space | Showing action and location together |
| Full shot | Subject head to toe, little background | Full-body movement, dance, sports |
| Medium shot | Waist up | Dialogue, everyday action, product use |
| Close-up | Head and shoulders | Emotion, reaction, product detail |
| Extreme close-up | Eyes, hands, or a small detail fills the frame | Tension, texture, tiny mechanisms |
| Over-the-shoulder | Foreground shoulder/head with subject beyond | Conversations, POV of a listener |
| POV | Camera IS the subject's eyes | Immersive first-person action |
Camera moves
Use exactly one move per shot — stacking moves (e.g. a pan that also cranes and dollies) confuses Sora and produces shaky, unpredictable motion.
| Move | Effect | When to use |
|---|---|---|
| Static / locked-off | No camera motion at all | Dialogue, calm product shots, tension through stillness |
| Pan | Camera pivots horizontally from a fixed point | Revealing a wide space or following lateral movement |
| Tilt | Camera pivots vertically from a fixed point | Revealing height — a building, a tall subject |
| Dolly in / out | Camera physically moves toward or away from subject | Building intensity (in) or releasing it (out) |
| Tracking / trucking | Camera moves alongside a moving subject | Following a walk, run, or vehicle at a constant distance |
| Crane / jib | Camera rises or descends smoothly, often changing angle | Grand reveals, rising above a scene |
| Arc / orbit | Camera circles around the subject | Hero shots, showing a subject from all sides |
| Push-in | Slow, subtle move closer without a full dolly | Quiet emotional emphasis |
| Pull-back | Slow move away to reveal context | Ending a scene, revealing scale or isolation |
| Whip pan | Very fast horizontal pan, motion-blurred | Energetic transitions, comedic or action beats |
| Handheld | Slight natural shake, human-operated feel | Documentary realism, urgency, chaos |
| Steadicam | Smooth floating motion while moving through space | Walking shots that need polish, not shake |
Lighting & lens
Pair one lighting term with one lens/style term for a consistent, filmic look.
| Term | Look |
|---|---|
| Golden hour | Warm, low-angle sun, long soft shadows |
| Blue hour | Cool twilight tones just after sunset |
| High-key | Bright, even, low-contrast lighting — commercial, upbeat |
| Low-key | Dark, high-contrast, moody shadows |
| Softbox | Diffused, flattering studio light |
| Practical lights | Light sources visible in-scene — lamps, neon, candles |
| Volumetric light | Visible light beams/haze — fog, dust, god rays |
| Backlight / rim light | Light behind subject, glowing edge outline |
| Shallow depth of field | Sharp subject, soft blurred background |
| 35mm | Natural, slightly wide documentary feel |
| 85mm | Compressed, flattering portrait feel |
| Anamorphic | Widescreen look with horizontal lens flares |
| Film grain | Subtle texture, analog/vintage feel |
Output settings
These are chosen in the Sora app or API before you generate — not written into the prompt text. Sora has no inline parameters like Midjourney's --ar.
| Setting | Options | Note |
|---|---|---|
| Duration | 4s, 8s, 12s, 16s, 20s (Pro up to ~25s via storyboard) | Shorter clips (4-8s) follow instructions more reliably |
| Resolution | 720p, 1080p, true_1080p | Draft at 720p, render the final pass higher |
| Aspect ratio | 16:9 (default), 9:16 (vertical) | 16:9 for YouTube, 9:16 for TikTok/Reels/Shorts |
| Audio | Synced dialogue + ambient sound + SFX | Generated automatically — describe what the scene sounds like |
Example prompts
Five complete shots built from the tables above, each ending with a plain-English settings note.
1. Product close-up
A ceramic coffee cup sits on a wooden counter, steam curling upward as a hand slowly rotates it. Kitchen, early morning light through a window. Close-up, static locked-off camera. Golden hour lighting, shallow depth of field, 85mm. Soft ambient kitchen sounds, faint ceramic clink as the cup turns.Best for: product shots and ads. (16:9, 1080p, 8s)
2. Walking city scene
A woman in a red coat walks briskly down a rain-slicked city sidewalk at night, weaving between umbrellas. Downtown street, neon signs reflecting in puddles. Wide shot, steadicam tracking alongside her. Low-key lighting, practical neon lights, film grain. Ambient traffic hum, rain patter, footsteps on wet pavement.Why it works: one action, one move, lighting matched to the mood. (9:16, 1080p, 12s)
3. Nature establishing shot
A lone hiker crosses a ridgeline as morning fog drifts through the valley below. Mountain range at sunrise. Extreme wide shot, slow crane rising to reveal the full valley. Golden hour lighting, volumetric light through the fog, 35mm. Wind gusting, distant birdsong.Best for: openers and travel content. (16:9, true_1080p, 12s)
4. Dialogue scene
Two friends sit across a café table, one leaning in to laugh at something the other just said. Cozy café interior, afternoon light through a large window. Over-the-shoulder shot, static camera. High-key lighting, softbox fill, shallow depth of field. Overlapping café chatter, espresso machine hiss, warm laughter and dialogue.Why it works: Sora's synced audio carries the scene as much as the visuals. (16:9, 1080p, 8s)
5. Action reveal
A vintage motorcycle roars out of a tunnel and onto an open desert highway at dusk. Empty desert road, blue hour sky fading to orange at the horizon. Full shot, dolly out as the bike accelerates toward camera. Backlight from the setting sun, anamorphic flares, film grain. Engine roar building, wind noise, gravel kicking up.Best for: high-energy hero shots. (16:9, 1080p, 8s)
Want more finished examples before you start? Browse the 35 best Sora prompts or grab ready-made skeletons from Sora prompt templates.
Frequently Asked Questions
How long can a Sora clip be?
Sora 2 clips run 4, 8, 12, 16, or 20 seconds. Sora 2 Pro can extend to roughly 25 seconds using storyboard mode. Shorter durations (4-8s) follow instructions more reliably than longer ones.
What resolutions does Sora support?
720p, 1080p, and true_1080p. Higher resolutions cost more generation time and credits, so many creators draft at 720p and re-render the final cut at 1080p or true_1080p.
Should I use 16:9 or 9:16?
Use 16:9 (the default) for YouTube and widescreen edits. Use 9:16 vertical for TikTok, Reels, and Shorts. Pick the aspect ratio in the app or API settings, not in the prompt text.
Can I set clip length in the prompt text?
No. Sora has no in-prompt parameters like Midjourney's --ar. Duration, resolution, and aspect ratio are chosen as app or API settings before you generate. Writing "10 seconds" inside the prompt is ignored.
Does Sora generate sound?
Yes. Sora 2 generates synced audio automatically, including dialogue, ambient sound, and sound effects. Describe what the scene should sound like in the prompt for better results.
How many camera moves should one shot have?
One. Stacking multiple camera moves (like a pan that also dollies and cranes) confuses Sora and produces shaky, unpredictable results. Pick a single move per shot from the camera moves table.