Every prompt here is written for a vertical feed and generates in native 9:16 — the single biggest thing that separates Shorts and Reels clips that look built for the format from cropped-down leftovers. They lean on Veo 3.1's native synced audio, so hooks get a spoken line, satisfying clips get crisp SFX, and avatars get real lip-sync. Paste any one into the Gemini app, Flow, or the Gemini API, set 9:16 in the settings, and generate. For the full library, see our 35 best Veo prompts roundup.
Scroll-stopping hooks
The first second decides everything on a vertical feed. These four open on motion or a spoken hook that stops the thumb, and each is composed for the tall canvas so nothing important sits near the cropped edges.
1. Direct-address opener
A young woman in a mustard hoodie snaps her head toward the lens and points straight at camera, eyes wide, mid-sentence. Sunlit bedroom with a plant shelf behind her. Vertical medium close-up framed for 9:16, subtle push-in. Bright natural window light, clean and crisp. She says, "Stop scrolling — you're doing this wrong." SFX: soft fabric rustle as she turns. Ambient: quiet room tone. (no subtitles, no on-screen text)Best for: a spoken hook that grabs the feed in the first half-second.
(9:16, 1080p, 6s)
2. The reveal drop
An open cardboard box sits center-frame on a white desk; two hands rip the tape and the flaps burst open toward camera as light spills out from inside. Bright studio tabletop. Vertical top-down-to-eye-level tilt, single move. High-key even lighting, glossy and clean. SFX: sharp tape rip, cardboard flex, a bright whoosh as the box opens. Ambient: soft studio room tone. (no subtitles, no on-screen text)Best for: unboxing and "wait for it" reveals that hold attention.
(9:16, 1080p, 6s)
3. Freeze-and-question hook
A man mid-stride on a city crosswalk suddenly freezes, one foot lifted, as everything around him keeps moving in a blur. Busy downtown intersection, overcast daylight. Vertical full shot framed for 9:16, slow arc around him. Flat diffused light, muted city palette. He says, "Ever wonder why nobody stops?" SFX: traffic swelling then dropping to silence around him. Ambient: distant crowd murmur. (no subtitles, no on-screen text)Best for: a curiosity-gap hook that sets up an explainer Short.
(9:16, 1080p, 8s)
4. Fast countdown tease
A pair of hands rapidly lays three objects onto a dark slate surface one after another — a phone, a coffee cup, a set of keys — each landing with a sharp beat. Moody tabletop set. Vertical top-down close-up for 9:16, static locked-off camera. Low-key lighting with a single hard rim highlight. SFX: three crisp thuds in rhythm, a final metallic key jingle. Ambient: low sub-bass hum. (no subtitles, no on-screen text)Best for: "3 things you need" list intros with a punchy rhythm.
(9:16, 4K, 6s)
POV & first-person
POV clips feel native to Shorts and TikTok because the camera is the viewer. Keep one clear action and let the audio sell the immersion. If you're building a recurring first-person series, upload reference images with Ingredients to Video to keep hands, gear, and setting consistent.
5. POV morning coffee
First-person POV: your own hands pour steaming coffee from a moka pot into a white mug on a wooden kitchen counter, morning light slanting across the steam. Cozy kitchen at dawn. Vertical POV framing for 9:16, gentle handheld sway. Warm golden window light, soft haze. SFX: coffee glugging into the mug, a soft ceramic set-down. Ambient: quiet kitchen, faint birdsong outside. (no subtitles, no on-screen text)Best for: aesthetic "day in my life" and routine openers.
(9:16, 1080p, 8s)
6. POV opening a door to somewhere epic
First-person POV: your hand pushes open a heavy wooden door and a sunlit mountain valley opens up in front of you, wind hitting your face. Rustic cabin threshold looking out. Vertical POV for 9:16, single slow dolly forward through the doorway. Bright natural daylight, high dynamic range. SFX: door creak, a swell of wind, a distant echo. Ambient: gusting mountain air. (no subtitles, no on-screen text)Best for: "wait for the reveal" travel and escape fantasy clips.
(9:16, 4K, 8s)
7. POV cooking close-up
First-person POV looking down at your hands sliding sizzling garlic and chili into a hot pan, tossing them with a flick of the wrist. Home stovetop, warm kitchen. Vertical overhead POV for 9:16, static locked-off camera. Warm practical kitchen light, glistening oil highlights. SFX: loud sizzle, oil crackle, the scrape of a wooden spoon. Ambient: gentle stovetop hum. (no subtitles, no on-screen text)Best for: fast recipe Shorts where the sizzle is the hook.
(9:16, 1080p, 6s)
8. POV skate roll
First-person POV low to the ground, rolling on a skateboard down a smooth concrete ledge as painted lines rush past underneath. Sunny skate plaza. Vertical POV for 9:16, forward tracking motion. Hard midday sun, crisp shadows, saturated colors. SFX: urethane wheels humming on concrete, a sharp trucks grind as it rolls off the edge. Ambient: distant park chatter. (no subtitles, no on-screen text)Best for: high-energy action POV that reads instantly on the feed.
(9:16, 1080p, 8s)
Satisfying & oddly satisfying
Oddly satisfying content thrives on clean motion and crisp sound. Describe the SFX precisely — the snap, the pour, the peel — because that audio is half the payoff. For a seamless loop, aim the motion back to its start and pair it with Frames to Video.
9. Perfect paint pour
Thick streams of teal, white, and gold paint pour onto a spinning canvas and swirl into a smooth marble pattern that never quite stops moving. Clean white studio tabletop. Vertical top-down close-up for 9:16, static locked-off camera. Soft even high-key light, glossy wet sheen. SFX: smooth viscous pour, a soft wet blending sound. Ambient: quiet studio hum. (no subtitles, no on-screen text)Best for: a hypnotic loop that holds viewers through the replay.
(9:16, 4K, 8s)
10. Kinetic sand slice
A sharp blade slices cleanly through a smooth block of pastel kinetic sand, the two halves parting to reveal a crisp inner texture. Minimal white surface. Vertical macro close-up for 9:16, subtle push-in. Bright soft studio light, gentle shadows. SFX: a satisfying crisp crunch as the blade cuts, fine grains trickling. Ambient: near silence. (no subtitles, no on-screen text)Best for: ASMR-style slicing loops with a crunch payoff.
(9:16, 4K, 6s)
11. Domino cascade
A curving line of glossy colored dominoes topples in a smooth wave across a dark table, each tile clicking down in sequence toward the camera. Dark matte tabletop. Vertical low-angle tracking along the line for 9:16, single move. Moody low-key light with bright specular highlights on each tile. SFX: rapid crisp click-click-click cascade building in speed. Ambient: low room tone. (no subtitles, no on-screen text)Best for: a rhythmic chain-reaction Short that begs a rewatch.
(9:16, 1080p, 8s)
12. Espresso pour loop
A thin ribbon of dark espresso streams into a clear glass and layers over milk in a clean gradient, crema settling on top in a smooth swirl. Marble café counter. Vertical macro close-up for 9:16, static locked-off camera. Warm side light, glossy glass reflections. SFX: espresso trickling, a soft liquid glug, faint crema fizz. Ambient: low café hum. (no subtitles, no on-screen text)Best for: café and barista loops that pair with a trending beat.
(9:16, 4K, 6s)
Talking avatar & voiceover
Veo 3.1 lip-syncs short spoken lines with real audio, so a talking-head Short needs only one or two sentences in quotes. Keep lines tight — a clip is 4-8 seconds. For a recurring host, lock the face with Ingredients to Video, and add the no-subtitles note so you can style your own captions.
13. Explainer host
A friendly presenter in a denim shirt talks straight to camera, gesturing with one hand to make a point, standing in a bright modern office. Vertical medium close-up for 9:16, static camera at eye level. Soft key light with a clean background blur. He says, "Here's the one setting everyone forgets." SFX: none. Ambient: quiet office room tone. (no subtitles, no on-screen text)Best for: the opening line of a tip or tutorial Short.
(9:16, 1080p, 6s)
14. Street interview avatar
A woman holding a foam mic leans toward a passerby on a busy sidewalk and asks a question, both smiling. Sunny city street, blurred crowd behind. Vertical medium shot for 9:16, gentle handheld sway. Bright daylight, natural color. She says, "Quick question — what would you do with an extra hour?" SFX: soft mic handling noise. Ambient: street chatter, distant traffic. (no subtitles, no on-screen text)Best for: faux street-interview formats that trend on Reels.
(9:16, 1080p, 8s)
15. Confessional voiceover
A person sits on the edge of a bed in a dim bedroom, lit only by a phone screen, speaking quietly and directly to camera. Late night, warm lamp glow off-frame. Vertical close-up for 9:16, very slow push-in. Low-key moody light, soft shadows. She says, "Nobody tells you this part." SFX: faint room hum. Ambient: quiet night, distant traffic. (no subtitles, no on-screen text)Best for: intimate storytime hooks that keep viewers watching.
(9:16, 1080p, 8s)
16. Product-in-hand pitch
A creator holds a small skincare bottle up beside her face and speaks to camera with a knowing smile, turning the bottle so the label faces the lens. Bright bathroom vanity with soft bulbs. Vertical medium close-up for 9:16, static camera. Flattering soft key light, clean whites. She says, "Okay, this one actually did something." SFX: soft cap click. Ambient: quiet room tone. (no subtitles, no on-screen text)Best for: UGC-style product talking heads; pairs with Veo product & ad prompts.
(9:16, 1080p, 6s)
Transitions & effects
Transitions carry a Short from one beat to the next without a cut. Build the move into the prompt — a whip, a wipe past an object, a spin — and let Veo carry the camera through it. Chain two clips with Extend to keep a transition rolling into the next scene.
17. Whip-pan location change
A person in a gray coat stands on a quiet street, then the camera whips hard to the right in a motion blur and settles on the same person now standing on a sunny beach. Two matched locations. Vertical full shot for 9:16, single fast whip pan. Natural daylight in both, matched exposure. SFX: a sharp whoosh at the whip, then beach waves. Ambient: street tone giving way to surf and gulls. (no subtitles, no on-screen text)Best for: outfit or location swaps with a snappy transition.
(9:16, 1080p, 6s)
18. Object-wipe transition
A hand swipes a large green leaf across the entire frame from left to right; as it passes, the scene behind it changes from a bare desk to a fully decorated one. Bright tabletop. Vertical medium shot for 9:16, camera static while the leaf wipes across. Soft even studio light. SFX: a soft leafy swish as it crosses. Ambient: quiet room tone. (no subtitles, no on-screen text)Best for: before/after and glow-up reveals in a single clip.
(9:16, 1080p, 6s)
19. Spin-and-outfit-change
A young man spins once on the spot in a plain white room and finishes the turn wearing a completely different outfit, arms out. Seamless white cyclorama. Vertical full shot for 9:16, camera holds while he spins. Bright high-key light, no shadows. SFX: a quick fabric swoosh on the spin, a light bass thump on the landing. Ambient: clean studio silence. (no subtitles, no on-screen text)Best for: fashion and outfit-of-the-day transition trends.
(9:16, 1080p, 6s)
20. Match-cut through a jump
A kid jumps up on a trampoline and at the peak of the jump the scene match-cuts so they land splashing into a swimming pool. Two matched backyard settings. Vertical full shot for 9:16, camera follows the upward motion. Bright summer sun, saturated blues and greens. SFX: trampoline creak and boing, then a big water splash. Ambient: backyard summer tone, cicadas. (no subtitles, no on-screen text)Best for: playful match-cut edits that feel bigger than one shot.
(9:16, 1080p, 8s)
Trends & memes
Trend formats move fast, but the shapes repeat: an expressive close-up, an absurd juxtaposition, a relatable everyday moment. These four give you a base clip you can drop a trending audio over. Keep any spoken line short so it doesn't fight the music.
21. Deadpan reaction close-up
A person stares dead-eyed into the camera, blinks once slowly, then raises a single eyebrow with perfect comedic timing. Plain neutral wall behind. Vertical tight close-up for 9:16, static locked-off camera. Flat even light, no drama. SFX: a single subtle blink-adjacent tick for timing. Ambient: total quiet room tone. (no subtitles, no on-screen text)Best for: reaction memes you overlay with trending audio.
(9:16, 1080p, 4s)
22. Absurd everyday juxtaposition
A business person in a full suit calmly eats cereal at a desk that is floating in the middle of a calm ocean, unbothered. Open sea, clear sky. Vertical wide shot for 9:16, slow pull-back revealing the water all around. Bright natural daylight, gentle reflections. SFX: spoon clinking in the bowl, soft ocean lapping. Ambient: distant gulls, light wind. (no subtitles, no on-screen text)Best for: surreal meme clips that get shared for the absurdity.
(9:16, 1080p, 8s)
23. Relatable Monday morning
A tired person slaps their alarm off, drags themselves upright, and stares blankly at the wall with hair sticking up. Messy bedroom, gray morning light. Vertical medium shot for 9:16, static camera at bedside level. Flat overcast light, muted tones. SFX: a jarring alarm cut short, a heavy sigh, rustling sheets. Ambient: quiet dull room tone. (no subtitles, no on-screen text)Best for: "me every Monday" relatable clips with caption overlay space.
(9:16, 1080p, 6s)
24. Pet mini-drama
A small dog sits on a couch and slowly turns its head to give a dramatic side-eye to the camera as if it just heard something outrageous. Cozy living room. Vertical close-up for 9:16, subtle push-in on the dog's face. Warm soft lamp light. SFX: a single tiny huff, a couch cushion shift. Ambient: quiet home room tone. (no subtitles, no on-screen text)Best for: pet-POV comedy meant for a dramatic voiceover.
(9:16, 1080p, 4s)
Mix and match across sections to build a full Short — open with a hook, cut to a POV beat, land a satisfying loop, close on a talking avatar. For the wider collection, browse the 35 best Veo prompts, and for talking-head work see our Veo dialogue prompts.
Frequently Asked Questions
Should I generate 9:16 natively or crop a 16:9 clip?
Generate 9:16 natively. Veo 3.1 supports vertical 9:16 as a real output setting, so it frames and composes for the tall canvas instead of leaving you to crop a landscape clip and lose the edges. Cropping a 16:9 render almost always cuts off heads, hands, or the subject's action. Choose 9:16 in the app or API before you generate.
How do I stop Veo from burning captions into the video?
Add "(no subtitles, no on-screen text)" to the end of the prompt. Veo 3.1 sometimes renders subtitles or floating text when a prompt contains spoken dialogue. The negative instruction keeps the frame clean so you can add your own styled captions in the editor, where they are far more legible for muted-autoplay feeds.
How good is Veo 3.1 lip-sync for talking-avatar Shorts?
Very good for short lines. Veo 3.1 generates dialogue as synced audio, so a single sentence in quotation marks lands with matching lip movement. Keep spoken lines to one or two short sentences per clip — a 4-8s clip has no room for a paragraph, and long monologues drift out of sync. For a full talking head, generate multiple clips and cut between them.
What clip length works best for Shorts and Reels?
Generate 8-second clips and trim to taste. Veo 3.1 outputs 4, 6, or 8 seconds per generation. Eight seconds gives you room to land a hook and a payoff, and you can string clips together with Extend for a 15-30s Short. For a single scroll-stopping loop, a tight 6-second clip often reads cleaner.
When should I use Veo 3.1 Fast?
Use Veo 3.1 Fast for volume — drafting hooks, testing variations, and posting daily. It is quicker and cheaper, which matters when a Shorts channel needs many clips a week. Draft on Fast, pick the winners, then re-render the keepers on full Veo 3.1 for maximum sharpness and audio fidelity.
How do I keep a character or product consistent across several Shorts?
Use Ingredients to Video. Upload up to three reference images — a character's face, an outfit, or a product — and Veo 3.1 keeps them consistent across separate generations. This is how you build a recurring avatar or a repeatable product-demo series without the look drifting shot to shot.
Can Veo make a seamless loop for a satisfying Short?
Yes, with a little help. Ask for a motion that returns to its starting position — a rotation, a pour that refills, a slice that resets — and use Frames to Video with the same first and last frame to close the loop. Crisp, described SFX (a clean snap, a soft click) make the loop feel satisfying even before you add music.
Do I need to describe sound, or does Veo add it automatically?
Describe it. Veo 3.1 generates synced audio, but you get far better results when you name the SFX and ambience you want — a knife tap, rain on a window, a crowd gasp. Spell out spoken lines in quotation marks and list SFX and ambient sound explicitly, because guessing gives you generic audio that rarely fits the beat.