This library of Veo 3 prompts makes 4 to 8 second clips: product shots, short dialogue scenes, nature footage, animated and ASMR-style shots, vertical Reels and photos brought to life. Each one is written for Toybox AI's Video Creator, which is powered by Google Veo 3.1, and each works as a starting point in Google Flow or the Gemini API, where results vary from run to run. The dialogue and sound lines follow Google's audio format for Veo 3.1. Video Creator clips come with sound, and there's no separate setting for it, so these lines are how you ask for voices and effects there too. The last section's prompts need a photo of your own as the start frame, and every AI video prompt here is ready to copy and paste, with a note on what to change.
How a good Veo 3 prompt is built
A good Veo 3 prompt describes one shot the way a director would brief a camera crew. Google's guides for Veo 3 and Veo 3.1 break it into these parts:
- Shot and camera: the framing and the move, such as "slow dolly in", "aerial view" or "low-angle tracking shot".
- Subject: who or what the shot is about, with specific details ("a man in his sixties with a short white beard and a green wool cap").
- Action: one thing that happens in the time you have.
- Setting: where it happens and what's around it.
- Light and style: the look, such as "golden hour", "35mm film" or "stop-motion clay animation".
- Sound: spoken lines in quotes after the speaker, "SFX:" for effects and "Ambient noise:" for the background.
Google's Veo 3.1 guide puts camera work first, then the subject, the action and the context, and ends with style and mood. Use this fill-in template for Veo 3 prompts as the base for your own:
[Shot type and camera move] of [subject with two specific details], [one action] in [setting with one or two details]. [Light], [style or film look]. [Speaker] says, "[one short line]". SFX: [one or two sounds]. Ambient noise: [background sound].For how each part works and how long a line can run in 8 seconds, read the Veo 3 prompting guide.
AI video prompt examples for products
These six show a product or a small business. The list of Veo 3 prompts in this section keeps each product in one place and lets the camera, the light or a pair of hands do the moving.
Coffee pour close-up
A pour shot for a cafe menu board or a social post.
Macro close-up, slow dolly in on a white ceramic mug on a wooden counter as dark coffee is poured from a copper kettle. Steam curls up through a beam of morning light from a side window. Shallow depth of field, warm tones, clean commercial look. SFX: coffee pouring, a soft clink as the kettle is set down. Ambient noise: a quiet cafe murmur.Change: the mug and drink, for example a glass of iced tea with lemon.
Candle lighting on linen
A slow, warm shot for a candle or home goods shop.
Tabletop product shot on a cream linen cloth. A hand strikes a match and lights a soy candle in an amber glass jar, and the flame catches and flickers. The camera orbits slowly halfway around the jar. Soft evening lamp light, cozy mood, shallow depth of field. SFX: a match strike and a faint crackle from the wick. Ambient noise: a quiet room.Change: the jar color and the surface, such as a marble tray or a wooden shelf.
Shoe on a turntable
A clean studio spin for a product page.
Plain white running shoe with no logos on a slowly rotating turntable against a light gray seamless backdrop. The shoe turns one full circle over the clip. Two softboxes light it from the left and right, with a soft shadow under the sole. Crisp commercial studio look, camera locked off at eye level. Ambient noise: a quiet studio with a faint hum.Change: the product and backdrop color, keeping the locked-off camera.
Ramen served from above
A top-down food shot for a restaurant.
Top-down shot of a dark wooden table. Hands set a steaming bowl of ramen in the center, then lay chopsticks across the rim as sliced green onions drift onto the broth. Warm restaurant lighting, rich colors, steam catching the light. SFX: a ceramic bowl set on wood, chopsticks tapping the rim. Ambient noise: a busy kitchen in the background.Change: the dish, and the hand action that serves it.
Walkthrough of a living room
A real estate style move through one room.
Smooth gimbal tracking shot moving forward through the front door into a bright living room with a gray sofa, oak floors and tall windows, then turning slowly toward an open kitchen with a white island. Midday natural light, airy and clean, real estate video style. Ambient noise: quiet room tone and birds faintly outside.Change: the room and the direction of the turn. Show only rooms that look like the real property.
Food truck in three shots
Three shots in one 8-second clip, using the timestamp format from Google's Veo 3.1 guide.
[00:00-00:03] Wide shot of a food truck parked on a sunny street corner at lunchtime, with a short line of customers. [00:03-00:06] Close-up of the cook flipping a tortilla on the griddle as steam rises. SFX: sizzling. [00:06-00:08] Medium shot of a customer taking the first bite and nodding slowly. Ambient noise: street chatter.Change: the business, keeping one establishing shot, one detail shot and one reaction.
Veo 3 prompt examples with dialogue
Dialogue prompts name the speaker by how they look, then give the line in quotes. Keep each line short enough to say comfortably in the clip.
Two clay mice split a crumb
A comic two-shot in stop-motion.
Stop-motion clay animation. Two clay mice sit at a tiny kitchen table with one crumb of cheese between them. The gray mouse says, "We split it evenly this time." The brown mouse slides the crumb toward itself and says, "Evenly is a big word." Warm miniature set lighting, medium two-shot, slightly jerky stop-motion movement.Change: the two animals and the object they argue over.
Cafe owner at opening time
A talking-to-camera clip for a local business.
Medium shot at eye level. A cafe owner in her fifties with short gray hair and a flour-dusted apron stands behind the counter, looks into the camera and says with a warm smile, "We open at seven, and the first cinnamon rolls come out warm." Soft morning light through the front window, handheld documentary style. Ambient noise: an espresso machine hissing.Change: the owner, the business and the line. Keep the line under 15 words.
Detective under a streetlamp
A moody film noir line.
1940s black-and-white film noir. Close-up of a tired detective in a wet trench coat under a flickering streetlamp at night. He tips his hat lower and says quietly, "The rain never tells you who it's hiding." Rain falls through the light, deep shadows, high contrast. SFX: steady rain on pavement and a distant car horn.Change: the era and genre, such as a 1970s crime drama in faded color.
Coach in the park
A short instruction line for a fitness page.
Medium wide shot in a sunny park. A fitness coach in a navy tank top holds a plank beside a class of three people on yoga mats, turns her head to them and says, "Ten more seconds. Keep your back straight." Bright morning light, green trees in soft focus, energetic handheld camera. Ambient noise: birds and a light breeze.Change: the exercise and the count in the line.
Cinematic Veo 3 prompts with ambient sound
These four have no dialogue, so the sound comes from the place itself.
Mountain lake at sunrise
A wide aerial opener.
Aerial drone shot gliding low over a still mountain lake at sunrise, with mist lifting off the water and pine forest on both shores. The camera rises slowly to reveal snowy peaks behind the lake. Golden hour light, cinematic, deep focus. Ambient noise: loons calling across the water and a soft wind.Change: the landscape, such as a coastline with sea stacks.
Cyclist in the rain
A low-angle night chase.
Low-angle tracking shot following a cyclist in a yellow raincoat through a narrow city street at night. Neon signs reflect in puddles, and the reflections ripple as the tires cut through them. Moody cinematic lighting, shallow depth of field, light rain falling. SFX: tires splashing, rain on shop awnings, a bicycle bell.Change: the rider's coat color and the city, such as a quiet old town with lanterns.
Dew drop on a fern
A macro shot with a focus pull.
Extreme close-up macro shot of a dew drop sliding down a fern leaf in a misty forest. The focus shifts from the drop to a small snail on a branch in the background. Soft, diffused green light, very shallow depth of field. Ambient noise: dripping water and distant birdsong.Change: the plant and the small creature in the background.
Storm over the prairie
A static wide shot where the weather does the moving.
Static wide shot of an open prairie under a building thunderstorm. Dark clouds roll toward the camera, lightning flickers inside them, and tall grass bends in the rising wind. Late afternoon light turning dim and green. Cinematic, deep focus, camera locked off on a tripod. SFX: a long roll of thunder. Ambient noise: wind through the grass.Change: the weather, such as a snowfall slowly covering a quiet farm road.
Animated and ASMR-style Veo 3 prompts
These four lean on a named style, a single sound or a timed sequence. Google's DeepMind guide suggests opening with the type of video you want, which the first three do.
Paper cutout fox
A handmade-looking animation for a children's story or a greeting.
Paper cutout stop-motion animation. A small orange paper fox with folded ears hops across layered paper hills under a paper moon, while a paper cloud slides slowly past behind it. Visible paper texture and soft shadows between the layers, warm handmade look, side view, camera locked off. SFX: paper rustling with each hop. Ambient noise: a quiet night with crickets.Change: the animal and the scenery layers, such as a paper whale under paper waves.
Watercolor harbor at dawn
A painted opener for a travel or story video.
Hand-painted watercolor animation of a small fishing harbor at dawn. Soft washes of peach and pale blue bleed at the edges, a wooden rowboat rocks gently at the dock, and two gulls glide across the sky. Slow pan from left to right, visible paper grain and loose brush strokes. Ambient noise: gentle waves against the dock and distant gulls.Change: the place and the hour, such as a snowy village at dusk.
Glass fruit ASMR slice
A satisfying close-up built around one sound.
ASMR macro close-up on a gray marble countertop. A chef's knife slowly slices through a translucent glass kiwi, and the slice tips over to show a clear green center with tiny glass seeds. Soft studio light, very shallow depth of field, camera locked off. SFX: a crisp glassy crunch as the blade passes through, then a light clink as the slice lands on the marble.Change: the object, such as a glass peach or a block of honeycomb.
Umbrella at a rainy bus stop
Three beats and one spoken line, using the timestamp format.
[00:00-00:03] Medium shot at a rainy bus stop at night. A young man in a gray hoodie holds a folded newspaper over his head as rain pours down. [00:03-00:06] A woman in a yellow raincoat steps under the shelter, opens a large umbrella and says, "Room for one more." [00:06-00:08] Close-up of him smiling as he ducks under it. SFX: heavy rain drumming on the shelter roof. Ambient noise: cars passing on wet pavement.Change: the weather and the line, keeping one setup shot, one line and one reaction.
Vertical Veo 3 prompts for Reels and Shorts
Set the format to vertical (9:16) before you run these three. Each keeps the subject centered, with room above and below it.
Guacamole in 8 seconds
A fast recipe clip.
Vertical overhead shot of a wooden cutting board. Hands halve a ripe avocado, scoop it into a clay bowl, mash it with a fork and squeeze half a lime over it. Bright kitchen daylight, fast social video pacing, everything centered. SFX: a knife on wood, the fork mashing, a quick lime squeeze.Change: the recipe, keeping three or four quick hand actions.
Terrier with a tennis ball
A dog clip at dog height.
Vertical low-angle shot at dog height on a sunny backyard lawn. A scruffy brown terrier races toward the camera with a tennis ball in its mouth, skids to a stop and drops the ball right in front of the lens. Bright afternoon light, playful handheld feel. SFX: paws on grass, excited panting, the ball bouncing once.Change: the breed, the toy and the yard.
Dress twirl on a rooftop
A simple outfit shot.
Vertical full-length shot on a city rooftop at sunset. A woman in a flowing red dress turns in a slow circle, and the fabric swings out around her. The camera stays still at waist height. Warm golden backlight, soft lens glow, city skyline blurred behind her. Ambient noise: a light breeze and faint traffic far below.Change: the outfit and the backdrop, such as a garden path in spring.
Veo 3 prompts that start from a photo
Upload your own image as the start frame, then describe only the motion. Match the photo's shape to the video: a vertical photo for a vertical clip, a wide photo for a wide one. Use only photos you're allowed to use, and read Toybox's privacy policy before uploading pictures of people.
Product photo with moving light
For a product photo you already have.
Starting from the uploaded photo, the camera pushes in slowly toward the product while a soft band of light moves across the background from left to right. The product stays still in the center of the frame, with its colors and shape unchanged. Calm, premium commercial mood.Change: the light direction, or swap the push-in for a slow pan.
Portrait that comes to life
For a portrait of someone who agreed to it.
Animate the uploaded portrait. The person blinks, smiles slightly and turns their head a little toward the window. Their hair moves gently in a light breeze. The camera stays still and the background stays softly blurred, matching the light in the photo.Change: the small action, such as a nod or a laugh. Keep it to one or two movements.
Landscape photo with drifting clouds
For a travel or scenery photo.
Bring the uploaded landscape photo to life. Clouds drift slowly from left to right, tall grass sways in the wind, and the camera pans slowly to the right. Keep the horizon and the colors of the original photo. Ambient noise: wind and distant birds.Change: the elements that move, such as water flowing in a river or leaves falling.
Tips for better Veo 3 results
Most of these come from Google's own Veo guides; the first is about working through this library:
- Change one line at a time. Swap only the detail in a prompt's "Change:" note, run it, and compare, so you know which edit changed the clip.
- One camera move per clip. A dolly, a pan or an orbit fills 8 seconds. Google's Veo 3.1 guide lists dolly, tracking, crane, aerial, slow pan and POV shots as camera moves to name.
- Tie every line to a speaker. Put each spoken line in quotation marks, as Google's API guide advises, and name the speaker by appearance ("the brown mouse") so it's clear who says it.
- Label the sound. Google's Veo 3.1 guide starts effect lines with "SFX:" and background lines with "Ambient noise:", and slips music into an SFX line. The guide to AI video with sound compares which tools return audio.
- Describe what's there, not what isn't. Google's example swaps a plain ban on man-made structures for "a desolate landscape with no buildings or roads". Say what should fill the frame.
- Write in English. English is the one language Google's API documentation lists as fully supported.
For anime-style clips, the Ghibli-style video prompts use the same structure with a hand-painted look.
Fixes for common problems
| Problem | Add this to your prompt |
|---|---|
| The camera drifts or shakes | "The camera is locked off on a tripod and doesn't move." |
| Two actions blur together | Split them with timestamps: "[00:00-00:04] ... [00:04-00:08] ..." |
| The wrong character speaks | Name the speaker by look: "The woman in the green coat says, ..." |
| The line sounds rushed | A shorter line, plus "She pauses, then says, ..." |
| Words appear on screen | "A clean frame with no on-screen text or captions." |
| A product label warps | "The product stays still in the center; only the light and background move." |
| The style drifts halfway | Start the prompt with the style: "Stop-motion clay animation." or "Realistic documentary footage." |
| The clip looks too dark | "Bright, even daylight" or "softbox lighting from the front." |
Run these prompts in Video Creator
Every "Try this prompt" link here drops the text into Video Creator. Pick the Format (Wide or Vertical) and the Length, attach your picture through "Start from a photo" for the last section's prompts, then tap "Create video". On Pro or higher, each clip currently takes 350 credits, and Video Creator Lite starts at 180. Clips come with sound, with no separate setting for it, so the dialogue and SFX lines shape what you hear. Spoken lines may not come out word for word, so listen to each clip before you post it. The Veo 3 explainer covers where the model came from and what it can do.