Veo Prompt Library

The Veo prompt library gives you copy-paste prompts for every major use case — product video, cinematic scenes, ASMR, vlog, image-to-video, dialogue, and more. Prompts are labeled by evidence tier: owner-tested, community-verified, or untested. Pick a use case below, copy the prompt, paste it into Veo, and adjust the details in bold.

This library covers Veo 4 (the latest release, up to 10–30 s, 4K, storyboard, avatar support) and Veo 3.1 / 3.1 Lite (free on all Google accounts). Where a prompt relies on a Veo 4-only capability — such as storyboard mode or extended duration — a note and a simplified fallback for Veo 3.1 are included alongside it.

Which Veo version should I use?

Use Veo 3.1 Lite if you want to experiment for free — it is available to every Google account in the AI Test Kitchen and VideoFX. Use Veo 3.1 (paid tier) for cleaner physics and longer outputs. Use Veo 4 when you need the highest fidelity, avatar generation, or durations beyond eight seconds; it requires access via Google AI Studio or Vertex AI. All prompts in this library work on Veo 3.1 unless otherwise noted; Veo 4-specific prompts are clearly labelled.

Don’t want to copy-paste and edit by hand?

The Veo Prompt Builder turns these same patterns into a structured form — pick a use case, fill in shot, lighting, and audio fields, and get a ready-to-run prompt in Text or JSON without hand-editing the bracketed placeholders above. It runs entirely in your browser, free, no signup.

Prompt deck

Copy a format, check the evidence, then customize it.

21 prompts 10 evidenced 10 community 0 owner-tested

Veo / All use cases

Storyboard-style three-beat ad (Coca-Cola JSON commercial)

Builder

Why it works The per-scene JSON gives Veo an explicit camera + sound cue for each 1.5–2s beat instead of one vague 8-second description, which is why the pacing, the cap-pop physics, and the final spoken brand line land cleanly. This is the same structure that makes storyboard-style ads reliable.

Prompt
{
  "video_length": 8,
  "scenes": [
    { "start": 0.0, "end": 2.0,
      "visual": "A cold Coca-Cola glass bottle stands upright against a deep red gradient background. It's covered in glistening condensation. The red bottle cap, embossed with the Coca-Cola logo, shines under a spotlight. Vapor gently rises from the base.",
      "camera": "quick dolly-in toward the bottle with a slight tilt up, shallow depth of field",
      "sound": "soft ambient fizzing, subtle whoosh as camera moves" },
    { "start": 2.0, "end": 3.5,
      "visual": "Close-up: the red Coca-Cola cap twists sharply and pops off with force. The cap spins in the air, showing the Coca-Cola logo in full as it rotates. Droplets fly off naturally with realistic gravity and inertia.",
      "camera": "snap zoom-in then slow-motion tracking of the cap mid-air",
      "sound": "crisp metallic twist, loud pop, carbonated hiss, followed by airy spin whoosh" },
    { "start": 3.5, "end": 5.5,
      "visual": "The Coca-Cola liquid flows out slightly, then wraps around the bottle in a high-speed swirl. The swirl follows a natural spiral pattern, with tiny droplets flying in all directions — rendered with realistic physics. The bottle remains still at the center.",
      "camera": "dynamic orbit shot around the bottle as liquid spins",
      "sound": "rich flowing liquid SFX, sparkling fizz buildup, airy rise" },
    { "start": 5.5, "end": 8.0,
      "visual": "Final wide shot: the bottle stands proud in the center. Red background glows subtly. Logo fades in above the bottle. A voice clearly says the brand name as the sonic sparkle finishes. Lens flare glides across as the screen fades out.",
      "camera": "locked hero shot, slow ambient glow increase",
      "sound": "bottle clink, soft chime, then voice saying the brand name with natural tone" }
  ]
}

TweakA timed, scene-by-scene JSON commercial. Swap the product name, background colour, and the spoken brand line in the final scene. The four beats — establish, open, swirl/action, logo — are the reusable pattern; keep the per-scene start/end timing.

Credit@aziz4ai, via songguoxs/awesome-video-prompts

Veo / All use cases

ASMR — glass-material slicing (viral hyper-real format)

Why it works A locked-off static shot removes camera motion as a variable, so all of Veo's effort goes into the slice physics and the shard glimmer — that focus is what makes the glass-cutting illusion and the implied tactile sound read as premium ASMR rather than a generic knife clip.

Prompt
Static shot, A man delicately slices a hyper-realistic glass dragon fruit on a pristine cutting board. The whisper-thin blade glides through the transparent fruit, scattering soft-glimmering shards. Surgical, serene lighting. Hyper-clean, ASMR video

TweakThe community "glass fruit" ASMR format that went viral. Swap "glass dragon fruit" for glass strawberry, glass onion, glass egg — "hyper-realistic glass [object]" + "static shot" + "ASMR video" is the load-bearing combination.

Credit@azed_ai, via songguoxs/awesome-video-prompts

Veo / All use cases

Stand-up comedian tells a joke (self-generated dialogue)

Builder

Why it works Framing the ask as a performance format ("stand-up comedy... tells a joke") rather than a scripted line lets Veo generate timing, delivery, and a punchline together — proof that native audio can carry a full comedic beat, not just a single quoted sentence.

Prompt
a man doing stand up comedy in a small venue tells a joke (include the joke in the dialogue)

TweakThe minimalist community prompt that first showed Veo writing AND delivering its own joke. To control the material, replace the parenthetical with your own line in quotes: says, "..."

Credit@fofrAI, via jax-explorer/awesome-veo3-videos

Veo / All use cases

Reference image character consistency (Google example)

Builder

Why it works The reference/"ingredients" images already fix the fish and the costume, so the short text prompt only has to specify the transformation and the motion — over-describing what the images already show is what makes reference-image prompts drift.

Prompt
Create a silly cartoon version of the fish wearing the costume, swimming and waving the wand around.

TweakUse when you have separate reference images for a character and an asset. Keep the prompt short: name the transformation and the motion, then let the reference images carry identity and costume details.

CreditGoogle AI for Developers — Veo documentation

Veo / All use cases

Beat-synced tech product reveal (Apple earbuds JSON)

Builder

Why it works Pairing a specific camera move with a specific sound cue per scene gives Veo a beat to cut to, which is what makes the flash-reveal and mid-air reassembly read as a tight, synced tech ad instead of a loose montage.

Prompt
{
  "video_length": 8,
  "scenes": [
    { "start": 0.0, "end": 0.7,
      "visual": "Apple earbuds appear in flashes over black void. Each flash reveals angle: top, side, front. Particles burst with light impact.",
      "camera": "snap zooms, hard cuts",
      "sound": "tight bass drops per cut" },
    { "start": 0.7, "end": 2.0,
      "visual": "Case pops open mid-air. Earbuds launch out in sync with beat, glowing rim light follows motion arcs.",
      "camera": "explosive transitions, 3D spin",
      "sound": "fast-paced pulse" },
    { "start": 2.0, "end": 3.5,
      "visual": "Earbuds split apart mid-flight. Internal parts float, orbiting like choreography.",
      "camera": "slow-motion breakaway",
      "sound": "digital glitch rhythm" },
    { "start": 3.5, "end": 5.0,
      "visual": "Floating parts twist and merge into Apple logo. Logo turns pitch black, neon rim lights glow softly.",
      "camera": "cinematic orbit + pull back",
      "sound": "echoing synth + Apple tone" },
    { "start": 5.0, "end": 8.0,
      "visual": "Apple logo holds center with ambient glow. Background fades to deep black. Silence.",
      "camera": "static frame",
      "sound": "quiet fade-out" }
  ]
}

TweakA beat-synced tech reveal: flash cuts, mid-air disassembly, then a logo merge. Swap the product and its logo; keep each scene under ~2 seconds so the cuts stay snappy and match a bass-drop rhythm.

Credit@aziz4ai, via songguoxs/awesome-video-prompts

Veo / All use cases

Instant-assembly unboxing reveal (IKEA bedroom JSON)

Builder

Why it works Listing every object that must appear gives Veo a concrete inventory to assemble instead of inventing generic furniture, which is why the room resolves into a specific, on-brand result rather than a vague "nice bedroom".

Prompt
{
  "description": "Cinematic shot of a sunlit Scandinavian bedroom. A sealed IKEA box trembles, opens, and flat pack furniture assembles rapidly into a serene, styled room highlighted by a yellow IKEA throw on the bed. No text.",
  "style": "cinematic",
  "camera": "fixed wide angle",
  "lighting": "natural warm with cool accents",
  "room": "Scandinavian bedroom",
  "elements": ["IKEA box (logo visible)", "bed with yellow throw", "bedside tables", "lamps", "wardrobe", "shelves", "mirror", "art", "rug", "curtains", "reading chair", "plants"],
  "motion": "box opens, furniture assembles precisely and rapidly",
  "ending": "calm, modern space with yellow IKEA accent",
  "text": "none",
  "keywords": ["16:9", "IKEA", "Scandinavian", "fast assembly", "no text", "warm & cool tones"]
}

TweakA "box explodes, room assembles itself" pattern for home/furniture brands. Swap the room type, brand box, and the explicit element list for your product line; keep the fixed wide angle so the assembly reads clearly.

Credit@Salmaaboukarr, via songguoxs/awesome-video-prompts

Veo / All use cases

Historical figure explains a concept to camera (Pythagoras)

Builder

Why it works Naming a well-known figure and a concept gives Veo enough context to generate period-appropriate speech, setting, and delivery without you writing a script — proof native audio can carry an entire explanatory monologue from a one-line brief.

Prompt
Pythagoras explaining his theorem, in ancient Greece

TweakA minimal prompt that still yields a full spoken explanation with period-accurate delivery. Swap the historical figure and the concept — Veo writes the explanation itself, so add a quoted line only if you need exact wording.

Credit@skirano, via jax-explorer/awesome-veo3-videos

Veo / All use cases

Cinematic tracking shot with a natural payoff beat (dachshund running)

Builder

Why it works The prompt tracks one continuous action through three connected spaces (room, doorway, porch) and ends on one specific, timed detail rather than a vague "and then something happens" — that specificity is what gives Veo a clear finish line for the camera move.

Prompt
The camera follows a dachshund running through a living room and out of an open front door and onto a porch. It stands on the top stair overlooking the neighborhood as an ice cream truck drives by.

TweakA clean example of a continuous tracking shot with a single scripted payoff detail (the ice cream truck). Swap the animal, the route, and the payoff detail at the end.

Credit@nmatares, via jax-explorer/awesome-veo3-videos

Veo / All use cases

Absurdist premise played straight (giraffe wheelie)

Builder

Why it works Stating the impossible action plainly and specifically ("pulls a wheelie", "streets of NYC") rather than adding hedging or extra adjectives lets Veo commit fully to the physical comedy instead of hedging the shot — useful evidence for any high-concept/attention-grabbing social clip.

Prompt
a giraffe pulls a wheelie on a dirt bike in the streets of NYC

TweakProof that a short, matter-of-fact sentence for an impossible premise still produces a coherent, physically grounded clip. Swap the animal, the vehicle, and the city for any other absurd-but-specific combination.

Credit@nmatares, via jax-explorer/awesome-veo3-videos

Veo / All use cases

Clean product hero orbit

Builder
Prompt
Premium commercial video of **a matte black wireless earbuds case** on a seamless light-grey studio surface. Slow 180-degree orbit, crisp edge highlights, shallow depth of field, high-key studio lighting. End on a centered hero frame with the product sharp. SFX: soft case click, subtle studio room tone. No text overlays, no watermark, no warped logo. 8 seconds, 16:9.

TweakBest for ecommerce hero shots. Swap the product and material; keep one camera move and one final hero frame.

Veo / All use cases

Category-set unboxing with a live reaction (Chewy pet supplies JSON)

Builder

Why it works Ending on a named reaction gives the assembly a payoff beat instead of stopping cold on the last object, which is what makes a multi-item unboxing feel complete rather than an inventory dump.

Prompt
{
  "description": "Cinematic shot of a sunlit, empty kitchen. A sealed Chewy box sits in the center. It trembles, explodes open in one burst, and pet supplies rapidly assemble into place: food and water bowls, a dog bed, toys, and a bag of food. A dog runs in and flops into the bed. No text.",
  "style": "cinematic",
  "camera": "fixed wide angle",
  "lighting": "natural warm with soft shadows",
  "room": "modern kitchen with hardwood floors",
  "elements": ["Chewy box (logo visible)", "dog food and water bowls", "dog bed", "dog toys (rope, ball, bone)", "bag of dog food", "wall hook with leash", "dog (golden retriever)"],
  "motion": "box explodes open, dog items fly out and assemble rapidly and precisely",
  "ending": "dog enters and settles happily into the bed",
  "text": "none",
  "keywords": ["16:9", "Chewy", "pet supplies", "fast assembly", "dog", "no text", "warm lighting"]
}

TweakThe unboxing pattern plus a payoff reaction (the dog settling in). Swap the product category and the "elements" list for your own set, and change the closing reaction to match who or what uses the product.

Credit@venturetwins, via songguoxs/awesome-video-prompts

Veo / All use cases

UGC testimonial with short spoken hook

Builder
Prompt
Vertical 9:16 selfie video of **a creator holding a skincare bottle** in a bright bedroom. Natural window light, slight handheld wobble, authentic phone-camera look. The creator smiles at the camera and says, "I did not expect this to work this fast." Casual upbeat tone, realistic lip sync, no captions, no text overlay. 8 seconds.

TweakKeep the spoken hook under 10 words. Long lines tend to become captions or rushed delivery.

Veo / All use cases

Dialogue scene with two short lines

Builder
Prompt
Medium two-shot in a quiet cafe. **A project manager** leans across the table and says, "We have one day to save the launch." **A designer** replies, "Then we stop polishing and ship the sharp version." Natural conversational timing, subtle background cafe ambience, no subtitles, no text overlay. 8 seconds, 16:9.

TweakTwo speakers work best when each line is short and clearly assigned.

Veo / All use cases

Image-to-video portrait animation

Prompt
The person in the photo breathes softly, blinks naturally, and turns their head a few degrees toward the camera with a faint smile. Hair and clothing move slightly as if a light breeze passes. Static framing, photoreal, natural motion. Ambient noise: soft outdoor air. No background music.

TweakUse with an uploaded portrait. Do not re-describe the face; let the source image carry identity.

Veo / All use cases

Cinematic reveal from a still image

Builder
Prompt
Begin tight on one detail in the input image, then slowly pull back and crane up to reveal the full scene and its scale. Steady continuous camera move, dramatic but smooth. Light blooms gently as the frame widens. Photoreal, cinematic, 8 seconds. Ambient sound matched to the scene.

TweakWorks for landscapes, architecture, product sets, and fantasy art. The image defines the scene; the prompt defines the reveal.

Veo / All use cases

Vertical street interview opener

Builder
Prompt
Vertical 9:16 handheld street interview at golden hour. The host steps into frame, points the microphone toward **a stranger in a denim jacket**, and asks, "What changed your mind this year?" Natural street ambience, casual documentary feel, slight handheld motion, no captions, no logo. 8 seconds.

TweakStrong for creator/social formats. One question only; save the answer for the next clip.

Veo / All use cases

Camera movement: slow dolly-in tension

Builder
Prompt
A quiet cinematic hallway at night, one door slightly open at the far end. Slow dolly-in from a medium-wide shot toward the door, low-key blue lighting, shallow depth of field, dust in the light beam. SFX: distant room tone and a faint floorboard creak. No jump scare, no text. 8 seconds.

TweakUse this as a reusable camera-control prompt. The dolly-in is the point; keep action minimal.

Veo / All use cases

No-subtitles talking-head control

Builder
Prompt
Close-up talking-head video of **a founder in a small studio** looking into the camera and saying, "The product is simple because the workflow is not." Calm, confident delivery, natural lip sync, soft key light, subtle room tone. Do not show subtitles, captions, lower thirds, logos, or any on-screen text. 8 seconds.

TweakUse the explicit negative line when Veo tries to print spoken words.

Veo / All use cases

Food product pour macro

Builder
Prompt
Extreme macro video of **dark chocolate sauce** pouring slowly over a scoop of vanilla ice cream. Low-angle camera glide, glossy highlights, shallow depth of field, appetizing commercial lighting. SFX: thick pour, soft plate contact, no music. No hands, no label, no text overlay. 8 seconds.

TweakA dependable food/F&B ad format. Replace the sauce and hero food; preserve the low-angle macro glide.

Veo / All use cases

Desk-lamp storyboard: problem, action, hero result

Builder
Prompt
Three-beat 8-second ad for **a compact desk lamp**. Beat 1: dim workspace at dusk, locked medium shot. Beat 2: hand taps the lamp, warm light blooms across the desk, slow push-in. Beat 3: hero close-up of the lamp beside a notebook, clean shadows, soft room tone. No text overlays, no watermark.

TweakThe simplest storyboard pattern: problem, action, hero result. Compare against the community JSON storyboard above if you want scene-by-scene timing control instead.

Veo / All use cases

Rainy-alley reference image character walk

Builder
Prompt
Using the supplied character reference image, show **the same character** walking through a rainy neon alley, pausing under a sign, then looking back toward the camera. Preserve face, outfit, color palette, and hair shape from the reference. Slow tracking shot, wet reflections, ambient rain and distant traffic. No face morphing, no outfit change. 8 seconds.

TweakUse with reference/ingredient images. Name what must stay consistent.