Vidu AI Video Generator: Turn Separate Photos into One Scene, Sound Included
What is Vidu AI?
Two Vidu models on Plimo that put the people, pets and places from your photos into one clip, with sound, at up to 4K.
Vidu AI is the video model from ShengShu Technology, a Beijing company that grew out of research at Tsinghua University. Its signature move is what ShengShu calls multiple-entity consistency: give it a portrait of your sister, a photo of her dog and a picture of the café where she works, and all three show up in one scene, each still looking like itself. Since Vidu 1.0 in April 2024, a new generation has landed every few months, from Vidu Q2 with sharper expressions to Vidu Q3 with picture and sound made together.
Plimo gives you two versions. Want a whole group from separate photos in one shot? Use Vidu Q4 Preview. Give it a portrait of each friend, even for a big group, and everyone ends up in the same frame, rendered up to 4K. Starting from a sentence, or from a then-and-now pair of pictures? Use Vidu Q3 Pro. It makes a scene from text alone, or glides from the old photo to the new one in one camera move. Both add sound and run up to 16 seconds.
Vidu also doesn't have to work alone. Kling, Veo, Seedance and Hailuo AI are in the same account, so when a Vidu take misses, the same photos go straight to the next model. The scene nobody could shoot together, made from the photos you already have.
Vidu Q4 Preview
Strong multi-subject reference: several people in one shot.
- 3–16 s
- 540P / 720P / 1080P / 2K / 4K
- Sound on or off
- Up to 15 reference photos
Vidu Q3 Pro
Text-to-video and first-and-last-frame video.
- 1–16 s
- 540P / 720P / 1080P
- Sound on or off
From Vidu 1.0 to Vidu Q4
April 2024
Vidu 1.0
ShengShu and Tsinghua University presented it as China's first large video model.
November 2024
Vidu 1.5
Brought in multiple-entity consistency: separate pictures of people, objects or places each stay recognizable inside one clip.
January 2025
Vidu 2.0
Went after speed and price. ShengShu said it cost half as much per second as Vidu 1.5.
April 2025
Vidu Q1
Could move between two unrelated pictures in one clip, and added background music and sound effects from text.
Late 2025
Vidu Q2
Sharpened facial expressions and camera control, and kept dialogue and effects in sync with the picture.
January 2026
Vidu Q3
Made picture and sound together in clips of up to 16 seconds, aimed at film and TV style scenes.
October 2026
Vidu Q4 Preview
An early look at Q4 that works from images only and fits a whole group from separate photos into one scene, with sound and resolutions up to 4K.
How to use Vidu on Plimo
Four steps from a handful of photos, or a single idea, to a Vidu clip with sound.
Upload your photos or write a prompt
Got separate photos of the people and things that belong in the shot? Upload them as reference images, one clear subject per photo. Only have an idea? Write it as a prompt. Just want one photo to move? Upload it as the first frame.
Pick Q4 Preview or Q3 Pro
Vidu Q4 Preview for group scenes built from separate photos, and for a 4K finish. Vidu Q3 Pro for a clip from words alone, or a move from one photo to another.
Say who does what
With reference photos, one sentence ties everyone together: “the dog leaps into the woman's arms on the beach.” With a single first frame, describe the motion, or leave the prompt empty and let Vidu choose one.
Generate and download
Choose a length of up to 16 seconds and generate. Sound comes with the picture, and you can switch it off when you plan to add your own music. Download the clip and post it.
Why use Vidu on Plimo
Vidu's group scenes and the rest of the newest video models, with one account and one set of credits.
- Highest resolution
- 4K
- Longest clip
- 16 sec
- Vidu versions
- 2
- Video models
- 15
A whole group, one scene
Vidu Q4 Preview builds one shot from a separate portrait of each friend, and everyone still looks like themselves. No other video model on Plimo fits a bigger group into one frame.Up to 4K, no watermark
Q4 Preview renders up to 4K, sharp enough for the TV at a family reunion. Every clip downloads clean, with no watermark on it.Clips from words alone on Q3 Pro
No photos yet? Vidu Q3 Pro builds the whole scene from a description, and can still start from your pictures when you have them.Sound in the same pass
Dialogue, ambience and effects come out with the picture on both versions. The clip is ready to share the moment it's done.Vidu without a ShengShu sign-up
Open Plimo in any browser and Vidu is there, on the account you already have. No second sign-up, no app to download.
Vidu compared with other video models
Kling
Want one long take of up to 30 seconds, or to edit a clip you filmed? Kling 4.0 is the better fit. Vidu is the pick when the cast comes from separate photos.
Veo
For a shot that looks filmed, Veo 3.1 Fast makes a polished 8-second clip with sound. Vidu lets you decide how long the clip runs.
Seedance
Want the moves copied from a short clip you shot? Pick Seedance. Vidu builds the scene from pictures alone.
Hailuo AI
If natural faces and expressions matter most, Hailuo AI's MiniMax H3 Max is known for them, with sound in every clip.
Grok Imagine
Want still images and videos from the same family? The Grok Imagine series makes both, and Grok Imagine 1.5 adds sound to every clip.
Wan
Want anything from a 2-second clip to a half-minute scene, or motion copied from footage you shot? Pick Wan 3.0.
FLUX
For a clip that runs past 16 seconds, FLUX 3 Video reaches 20, steered by your description and the pictures it starts and ends on.
SkyReels
Need a shot that lands exactly on your poster or final picture? SkyReels V4 builds toward that ending, though it always needs a written prompt.
HappyHorse
Want lively, expressive people with sound that lands on every beat? Pick HappyHorse 1.1, though it needs a prompt every time.
What people make with Vidu
Vidu shines when the people and things you want together were never in the same photo.
A group shot nobody was in together
Add a separate portrait of each friend and describe everyone crowding around one birthday cake as the candles go out. The whole group finally fits in one picture.Family across different cities
Use photos of grandparents, cousins and the kids, taken in three different places, and set them all around one dinner table. A reunion clip for the family chat, without the plane tickets.You, your dog and your favorite café
Use reference photos of your jacket, your dog and a café interior, then ask for you walking in wearing the jacket while the dog trots at your heels.Then and now in one move
On Vidu Q3 Pro, set a childhood photo as the first frame and a recent one as the end frame. The clip travels from one to the other in a single camera move.A product in a styled setting
Give Q4 Preview a product photo, a model shot and a backdrop, then describe the model lifting the item off a shelf and turning it toward the camera. Your listing shows the product in use, with no shoot to book.A recurring character for your channel
Reuse the same reference photos of a mascot or a host in every episode, and the character looks the same from one clip to the next.A scene from words alone
Vidu Q3 Pro doesn't need an image. Describe a paper lantern rising over a night market and drifting past lit rooftops, with the murmur of the crowd underneath.A portrait that picks its own motion
Give Q4 Preview a single portrait as the first frame, leave the prompt empty and see whether it chooses a smile, a turn of the head or a breeze through the hair.One long take with no cuts
Write a single continuous shot on Q3 Pro, like a cyclist weaving through a morning market from one end to the other, and let it run the full 16 seconds.A vertical clip for social media
Make it vertical and put yourself and your friends from separate selfies into one scene for TikTok, Reels or Shorts, sound included.