Vidu, from Shengshu Technology, turns a sentence, a still, or a handful of reference images into a short cinematic clip — and it follows camera and motion language unusually well. The 2026 lineup runs on the Vidu Q2 line: Q2, Q2 Pro for the highest detail, and Q2 Turbo for fast drafts, all with synced 48kHz audio. Its two real edges over rival video models are reference-to-video — up to seven reference images fused into one consistent clip — and clean anime motion.

Every prompt below is written the way Vidu rewards: subject first, then one action, then an explicit camera move, then lighting and mood. Under each is a Settings note — resolution, aspect ratio, length and movement amplitude are app controls in Vidu, not inline parameters, so set them there and swap the bracketed parts. Keep the Vidu prompt cheat sheet open for the camera-move and amplitude tables, and read how to prompt Vidu for realistic video for the formula behind the b-roll sections.

Advertisement

Cinematic b-roll

Establishing shots live and die on the camera move. Name one move per clip, set the light, keep the scene simple, and hold movement amplitude low so Vidu can render it steady.

1. Foggy valley at dawn

A wide mountain valley at dawn, layers of pine ridges fading into low fog, thin clouds drifting between the peaks, slow aerial drone push forward over the treetops, soft golden first light breaking on the horizon, cinematic color grade, calm and epic.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: small. Why it works: one slow aerial move plus drifting fog gives Vidu a single clear direction of motion, so the clip feels directed instead of jittery.

2. Neon night market

A crowded night market in a rainy Asian city, glowing red and blue signs reflected in wet pavement, steam rising from food stalls, slow dolly forward down the center of the lane, shallow depth of field, moody cinematic lighting, soft ambient market chatter and sizzling.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: auto · Background Music off. Best for: title backgrounds and moody establishers. The audio line triggers Vidu's synced ambience.

3. Coastal cliff drone reveal

A rugged coastline with turquoise waves crashing against tall cliffs, a lone lighthouse on the headland, aerial drone shot rising and pulling back to reveal the full coast, bright midday sun, deep blue ocean, crisp natural color, smooth cinematic motion.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: small. Why it works: a rise-and-pull-back is Vidu's strongest reveal move, and a single landmark anchors the eye as the frame widens.

4. Autumn forest path

A narrow path through an autumn forest carpeted in orange and gold leaves, a few leaves drifting down, sunlight filtering through the canopy in soft shafts, slow steady push-in along the path, warm golden light, peaceful and nostalgic, gentle depth of field.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: small. Best for: calm nature b-roll and background loops. Add "soft breeze and rustling leaves" for synced audio.

5. City skyline dusk time-lapse

A modern city skyline at dusk transitioning to night, office windows lighting up one by one, streaks of car headlights on the roads below, slow time-lapse with a gentle push-in toward the tallest tower, deep blue-to-orange gradient sky, cinematic and polished.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: auto. Why it works: "time-lapse" plus a slow push-in tells Vidu to animate the light change while keeping the camera calm.

6. Rain on a window

Close-up of raindrops running down a café window at night, warm blurred bokeh of street lights and signs beyond the glass, a single drop sliding and merging with others, static camera with a subtle slow push-in, moody warm-and-blue tones, quiet and intimate, soft rain sound.

Settings: Vidu Q2 · 9:16 · 1080p · 6s · amplitude: small. Best for: mood shots and vertical intros. A nearly locked camera keeps the bokeh clean.

People & motion

Describe the person, give them one action, and let the camera carry the story. Vidu holds a single figure in motion — and their facial expression — far better than a crowd.

7. Barista pouring latte art

A barista in an apron pouring steamed milk into a cup to form a leaf of latte art, steam curling upward, slow push-in on the cup and her hands, soft window light with a warm bounce, shallow focus, cozy documentary feel, calm and precise.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: small. Why it works: hands and pouring milk are small, contained motions, so a slow push-in keeps them sharp and readable.

8. Runner at sunrise

A young woman jogging along an empty seaside promenade at sunrise, breath visible in the cool air, ponytail bouncing, tracking side shot moving with her at a steady pace, warm low backlight rimming her silhouette, motivational sports feel, subtle lens flare.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: medium. Best for: fitness and lifestyle ads. Medium amplitude gives a livelier stride without wobble.

9. Potter at the wheel

A ceramicist's wet hands shaping a spinning clay bowl on a pottery wheel, the clay rising and narrowing under her fingers, water glistening, slow orbit around the wheel from the side to the front, warm natural window light, calm focused craft mood, fine detail on the clay.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: small. Why it works: a slow orbit adds depth while the subject's motion stays contained, which keeps the hands stable.

10. Dancer in an empty studio

A contemporary dancer in flowing fabric moving through a slow expressive turn in a bright empty studio, dust motes floating in a shaft of light, sweeping tracking shot arcing around her, soft directional daylight, elegant and emotional, gentle motion blur on the fabric.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: medium. Best for: expressive brand and music moments. One turn plus one arc keeps the choreography clean.

11. Child blowing a dandelion

A laughing child in a summer meadow blowing the seeds off a dandelion, seeds drifting and scattering into the warm air, handheld tracking shot easing in toward their face, bright golden afternoon light, joyful and candid, shallow focus, soft breeze sound.

Settings: Vidu Q2 · 9:16 · 1080p · 6s · amplitude: medium. Why it works: the drifting seeds give lively secondary motion while the child's face stays the calm anchor.

Advertisement

Product & commercial

Center the product, keep the background clean, and choose a controlled camera move — a slow rotation or a macro push reads as premium. Hold movement amplitude at small so the product never warps.

12. Perfume bottle rotation

A faceted glass perfume bottle with a gold cap on a wet black reflective surface, a soft swirl of pink mist drifting behind it, slow 360-degree rotation of the bottle, elegant studio lighting with a bright rim highlight on the edges, luxury commercial look, crisp reflections.

Settings: Vidu Q2 Pro · 1:1 · 1080p · 6s · amplitude: small. Why it works: a slow single-axis rotation on a clean set is the classic product-hero move, and low amplitude keeps the glass stable.

13. Sneaker splash

A white-and-orange running sneaker stepping into a shallow puddle in slow motion, water splashing up in a crown of droplets around the sole, tracking macro shot at ground level, punchy directional light, high-contrast commercial color, crisp frozen droplets, dark studio background.

Settings: Vidu Q2 Pro · 16:9 · 1080p · 5s · amplitude: medium. Best for: energetic footwear ads. "Slow motion" plus a ground-level track sells the splash.

14. Watch floating spin

A luxury chronograph watch floating and slowly rotating above an open minimalist box, soft particles of light drifting around it, clean gradient studio background from charcoal to deep blue, crisp rim lighting on the case and bracelet, modern premium tech-commercial feel.

Settings: Vidu Q2 Pro · 16:9 · 1080p · 6s · amplitude: small. Why it works: a floating slow spin on a gradient set is a reliable luxury look, and drifting light adds life without risking the product.

15. Serum droplet macro

A frosted-glass serum dropper releasing a single glossy golden droplet onto a smooth cream surface, the drop landing and rippling outward in slow motion, soft even beauty lighting, clean minimalist studio in sage and cream tones, macro focus, calm premium mood.

Settings: Vidu Q2 Pro · 1:1 · 1080p · 5s · amplitude: small. Best for: cosmetics and skincare hero clips. Small amplitude keeps the drip controlled.

16. Cola ice pour

Dark cola pouring over ice cubes into a frosted glass, bubbles fizzing and rising, condensation beading on the glass, static locked-off camera with a subtle slow push-in, bright appetizing side light, rich caramel tones, refreshing high-contrast commercial look, crisp fizz sound.

Settings: Vidu Q2 Pro · 9:16 · 1080p · 5s · amplitude: small. Why it works: a nearly locked camera lets the pour and fizz be the only motion, which keeps a macro shot from smearing.

Anime & stylized

Anime is Vidu's home turf. Name the exact style plus line weight and shading, keep one character in frame, and add a camera move for energy. Vidu can also handle a short multi-shot script in one prompt.

17. Anime girl on a rooftop

A young anime girl with short teal hair standing on a school rooftop at sunset, hair and skirt swaying in the wind, she turns her head toward the camera and smiles softly, slow push-in on her face, warm orange light and long shadows, wistful Makoto Shinkai-inspired style, crisp clean line art, detailed sky.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: auto. Why it works: the swaying hair supplies motion while her face stays calm — exactly what anime frames handle beautifully.

18. Ghibli-style meadow walk

A girl in a red dress walking through a rolling green meadow of wildflowers under a huge blue sky with fluffy clouds, grass and hair swaying in the breeze, slow tracking shot following beside her, soft hand-painted Studio Ghibli-inspired style, warm nostalgic light, gentle and dreamy.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: small. Best for: painterly openers. A side track plus swaying grass gives that wandering feel without complex character animation.

19. Cyberpunk courier ride

A cyberpunk courier on a glowing hover-bike weaving between neon skyscrapers at night, trails of light streaking past, dynamic tracking shot swooping alongside the bike, rain streaking through the headlights, vibrant cel-shaded anime style with bold outlines, high energy, motion blur.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: large. Why it works: large amplitude plus a clear alongside track keeps the chase fast and directed.

20. Cel-shaded sword draw

A lone samurai in a bamboo forest slowly drawing a katana, petals and mist drifting, the blade catching a glint of light, dramatic low-angle push-in on the draw, bold cel-shaded anime style with thick outlines and flat shading, tense and quiet, sharp rim light, soft wind sound.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: medium. Best for: action anime beats. The slow draw plus a single push-in builds tension cleanly.

21. Watercolor cat by a window

A fluffy grey cat curled on a windowsill in the rain, tail flicking slowly, ears twitching, soft loose watercolor illustration style with visible paper texture, muted blue and warm-cream palette, slow gentle push-in, calm and cozy, blurred rain beyond the glass.

Settings: Vidu Q2 · 1:1 · 1080p · 6s · amplitude: small. Why it works: naming the medium and a slow push-in keeps the painterly texture intact instead of resolving into photoreal fur.

22. Anime three-shot sequence

Anime scene, warm sunset color, clean line art. Shot 1: a boy stands on a train platform holding a letter, wind lifting his hair. Shot 2: close-up of his determined face as he looks up. Shot 3: he runs toward the departing train, cherry petals swirling behind him. Smooth cuts between shots, cinematic.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: medium. Why it works: Vidu handles a simple Shot 1 / Shot 2 / Shot 3 script, so you get a tiny storyboard in one clip. Keep each shot to one action.

Advertisement

Action & VFX

Give the effect a clear cause and a single camera move, and raise movement amplitude for impact. One dramatic beat per clip reads better than a stack of them.

23. Superhero rooftop landing

A caped superhero landing hard on a rain-slicked rooftop at night, cracking the concrete with a shockwave of dust and water, cape settling, low hero angle looking up as the camera slowly pushes in, dramatic blue moonlight with a warm city glow behind, cinematic blockbuster look.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: large. Why it works: the impact gives a burst of motion up front, then the slow push-in lets it settle into a hero beat.

24. Dragon over a valley

An enormous scaled dragon sweeping low over a misty green valley at dawn, wings beating, trees bending in the downdraft, sweeping aerial tracking shot flying alongside the dragon, mountains and a river beyond, epic fantasy color grade, cinematic scale, deep wind rumble.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: medium. Best for: fantasy trailers. A fly-alongside track keeps the creature centered and majestic.

25. Explosion walk-away

A rugged action hero in a leather jacket walking calmly toward the camera as a massive fireball erupts behind him, debris and sparks flying, he does not look back, slow-motion dolly back keeping pace with him, warm orange firelight on his face, gritty blockbuster grade.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: medium. Why it works: the don't-look-back setup gives Vidu two clear layers — steady subject, chaotic background — and a dolly-back to hold the frame.

26. Motorcycle tunnel ride

A sport motorcycle speeding through a neon-lit tunnel at night, light strips streaking past on the walls, rider leaning into the motion, dynamic tracking shot chasing just behind the bike, reflections racing across the wet asphalt, high-energy cinematic grade, motion blur, engine roar.

Settings: Vidu Q2 · 16:9 · 1080p · 6s · amplitude: large. Best for: high-adrenaline openers. A chase-behind track plus streaking light sells the speed.

27. Underwater whale pass

An enormous humpback whale gliding slowly through deep blue ocean water, shafts of sunlight streaming down from the surface, tiny fish scattering, slow tracking shot drifting alongside the whale, serene underwater light with floating particles, majestic and calm, cinematic.

Settings: Vidu Q2 · 16:9 · 1080p · 8s · amplitude: small. Why it works: the drifting-alongside track keeps the whale grand and unhurried, and low amplitude suits its slow glide.

Reference-to-video

This is Vidu's signature mode. Upload up to seven reference images — characters, props, or scenes — describe how they interact, and add an identity-lock line. See the full pack of Vidu reference-to-video prompts.

28. Character meets a dragon

Using the reference images, the young warrior from image 1 stands in the mountain landscape from image 2 and reaches out toward the small dragon from image 3, which flutters down to land on her arm, slow push-in, warm dawn light, cinematic fantasy, keep each character consistent with its reference image.

Settings: Vidu Q2 Reference-to-Video · 3 reference images · 16:9 · 1080p · 6s · amplitude: medium. Why it works: naming which image each element comes from plus a keep-consistent line lets Vidu fuse unrelated references into one coherent shot.

29. Mascot holds the product

Using the reference images, the brand mascot from image 1 holds up the product from image 2 and gestures to it with a cheerful smile, clean bright studio background, slow push-in on the product, upbeat commercial lighting, keep the mascot and the product exactly consistent with their reference images.

Settings: Vidu Q2 Reference-to-Video · 2 reference images · 1:1 · 1080p · 5s · amplitude: small. Best for: consistent branded ads where the character and packaging must match every time.

30. Two characters shake hands

Using the reference images, the businesswoman from image 1 and the businessman from image 2 meet in the modern office lobby from image 3, shake hands and smile, slow orbit around the two of them, soft daylight through tall windows, professional and warm, keep both characters consistent with their references.

Settings: Vidu Q2 Reference-to-Video · 3 reference images · 16:9 · 1080p · 6s · amplitude: medium. Why it works: a single shared action — the handshake — gives Vidu one clear interaction to solve between two locked identities.

31. Model wears the outfit

Using the reference images, the model from image 1 wears the jacket from image 2 and walks toward the camera down a city sidewalk, turning slightly to show the fit, tracking shot easing back with her, soft overcast fashion light, editorial mood, keep the model's face and the jacket exactly consistent with their references.

Settings: Vidu Q2 Reference-to-Video · 2 reference images · 9:16 · 1080p · 6s · amplitude: medium. Best for: fashion and try-on clips that keep both the person and the garment true.

32. Character in a new setting

Using the reference image, the character from the reference explores the neon cyberpunk street around them, looking around in wonder as signs flicker overhead, slow tracking shot following just behind, rain and reflections, moody cinematic light, keep the character's face, hair and outfit consistent with the reference image.

Settings: Vidu Q2 Reference-to-Video · 1 reference image · 16:9 · 1080p · 6s · amplitude: medium. Why it works: one locked character plus a described new world is the simplest reliable reference-to-video setup.

33. Product in a lifestyle scene

Using the reference images, the coffee bag from image 1 sits on the wooden kitchen counter from image 2 in soft morning light, steam rising from a filled mug beside it, hands entering to lift the mug, slow push-in, warm cozy lifestyle look, keep the product packaging exactly consistent with its reference.

Settings: Vidu Q2 Reference-to-Video · 2 reference images · 16:9 · 1080p · 6s · amplitude: small. Best for: placing a real product into a lifestyle shot with the label kept crisp and correct.

Advertisement

Image-to-video

These animate a still you upload in image-to-video mode. Describe only the motion you want added and lock the look — Vidu keeps the frame and moves it. The last two use start-and-end-frame mode. See the Vidu prompt templates for reusable skeletons.

34. Portrait comes to life

Bring this portrait to life: the person blinks naturally and gives a soft warm smile, hair moving slightly in a gentle breeze, subtle head turn toward the camera, keep the lighting, background and face identical to the photo, slow subtle push-in, natural and realistic.

Settings: Vidu Q2 Image-to-Video · 1080p · 5s · amplitude: small. Why it works: naming small, natural motions and saying "keep identical" gives a believable live portrait without distorting the face.

35. Landscape photo parallax

Animate this landscape photo with gentle parallax: clouds drift slowly across the sky, water ripples and reflects, grass and leaves sway in a light breeze, a slow cinematic push-in adds depth, keep every element exactly as in the image, calm and alive.

Settings: Vidu Q2 Image-to-Video · 1080p · 8s · amplitude: small. Best for: turning a photo into a living wallpaper. Assigning motion to layers — sky, water, foliage — creates convincing depth.

36. Old photo animated

Gently animate this vintage family photograph: the people breathe and blink softly, small natural head movements and warm smiles, faint film grain and a subtle slow zoom, preserve the original colors, clothing and composition exactly, nostalgic and heartfelt.

Settings: Vidu Q2 Image-to-Video · 1080p · 5s · amplitude: small. Best for: memory and heritage content. Low amplitude keeps likenesses true.

37. Logo particle reveal

Animate this logo: the mark assembles from glowing particles that fly in and settle into the final shape, a soft light sweep passes across it, gentle floating idle motion, clean modern brand animation, keep the exact colors and shape of the logo, dark gradient background.

Settings: Vidu Q2 Image-to-Video · 16:9 · 1080p · 5s · amplitude: medium. Best for: intros and stingers. The particle-assemble effect is a dependable Vidu look.

38. Start-to-end frame morph

Smoothly transition from the start frame to the end frame: the camera pushes in gently as the scene morphs, matching lighting and composition between the two frames, seamless cinematic transition, natural motion, no sudden jumps.

Settings: Vidu Q2 Start-End Frame · 16:9 · 1080p · 5s · amplitude: small (keep both frames within a 0.8–1.25 aspect ratio). Why it works: a clear "transition, keep lighting matched" instruction is all start-and-end mode needs — it anchors identity from both frames. For the full toolkit, keep the Vidu prompt cheat sheet handy.

Frequently Asked Questions

What is Vidu and who makes it?

Vidu is a multimodal AI video generator built by Shengshu Technology (ShengShu-AI). It turns text prompts, still images and multiple reference images into short cinematic clips. The 2026 lineup runs on the Vidu Q2 line — Q2, Q2 Pro for the highest detail, and Q2 Turbo for fast drafts. It is best known for two things: reference-to-video, which combines up to seven reference images into one cohesive clip with consistent characters, and clean anime and stylized motion. You can use it at vidu.com and through API partners.

How do I write a good Vidu prompt?

Vidu rewards clear structure. Name the subject and its appearance first, then one action or motion, then an explicit camera move, then the setting and lighting, and finish with the mood. Keep it to a single continuous action per clip and around 25 to 60 words. Vidu follows camera language closely, so name a real move such as slow push-in, pull-back, orbit or aerial tracking. State resolution, aspect ratio and length as settings in the app rather than inside the prompt.

What is reference-to-video in Vidu?

Reference-to-video lets you upload up to seven reference images — characters, props, or scenes, even unrelated ones — and Vidu blends them into a single coherent clip while keeping each subject looking like its reference. It is Vidu's signature feature and the reason it is used for consistent characters across shots. In the prompt, describe how the referenced subjects interact and add a line like keep each character consistent with its reference image. Save frequently used subjects in the My References library to reuse them.

What resolutions and lengths does Vidu support?

Vidu Q2 generates at 540p, 720p and 1080p at 24fps, with clip lengths of roughly 2 to 8 seconds per generation. Aspect ratios include 16:9 for YouTube, 9:16 for TikTok and Reels, and 1:1 for feed posts. For longer stories, generate several short shots and stitch them in an editor, or use the start-and-end-frame mode to bridge two frames smoothly.

What is movement amplitude in Vidu?

Movement amplitude controls how much motion Vidu adds, with settings of auto, small, medium and large. Auto is a safe default and works well for portraits and calm scenes. Use small for subtle, professional product shots and smooth transitions, and medium or large for full-body action, fast camera moves and dramatic effects. Pair a larger amplitude with a clear camera instruction so the extra motion has direction instead of wobble.

Does Vidu generate audio?

Yes. Vidu Q2 can generate 48kHz sound effects synced to the on-screen action, and a Background Music toggle adds a matching soundtrack at no extra credit or duration cost. Describe the sound you want in plain language inside the prompt, for example add gentle rain and distant thunder or soft ambient market chatter, and Vidu matches the audio to the motion. You can also leave a clip silent and add your own audio in an editor later.

Text-to-video, image-to-video or reference-to-video: which should I use?

Use text-to-video when you want Vidu to invent the whole scene from a description. Use image-to-video when you already have the exact look locked in one still and only want it to move, which gives the most control over composition. Use reference-to-video when you need specific characters, props or scenes to stay consistent, especially across multiple clips — this is Vidu's strongest and most distinctive mode. Start-and-end-frame is a fourth option that fills the motion between a first and last frame.

How long should a Vidu prompt be?

Aim for roughly 25 to 60 words for a single shot. That is enough to name the subject, the action, one camera move and the lighting without burying the model in competing instructions. Vidu also handles a short multi-shot script — Shot 1, Shot 2, Shot 3 — inside one prompt, which is useful for anime sequences, but keep each shot to one clear action and expect to stitch longer stories in post.

Advertisement