Anime that moves well
Vidu is pitched at animation, and its anime clips keep cel-shaded characters on model through fast action, like our skater's ollie.
Vidu · by Shengshu
Vidu AI video generator
Vidu is Shengshu's video model family. Vidu Q3 is known for anime and short-drama clips with sound and cuts inside one generation; Vidu Q4 brings a photo or a set of reference images to life with the same face. Write a scene below and it opens in Video Studio on Vidu Q3.
Example video
Opens in Video Studio with the model you picked. You see the price in credits before anything runs.
What Vidu is known for
Short drama with sound
Q3 generates dialogue, effects and music with the picture and cuts between shots inside one clip; every line in our gallery was spoken as written.
Your photo, your character (Q4)
Vidu Q4 animates a start frame or carries people, outfits and products over from reference images. Our gym still came to life with the same face and the same bottle label.
Long single clips
Vidu Q4 runs up to 16 seconds on Affogato, among the longest single clips in its class; Q3 runs 5 or 10 seconds in the studio; the 8-second Q3 clips below were made through Vidu's own API, which allows that length.
Where it's weaker
- Multi-person scenes are less consistent, and lip sync in dialogue lags the best models.
- Scripted reactions can go missing: our drama clip delivered both lines but not the laugh the prompt asked for.
- Crowds and complex physical contact are a weak spot.
- Vidu Q4 is a preview of the full model, and it has no text-only mode: it needs a photo or reference images.
- Maker
- Shengshu · Q3 January 2026, Q4 Preview October 2026
- Clip length on Affogato
- 5 or 10 s (Q3) · 4, 5, 8, 10 or 16 s (Q4)
- Resolution
- Up to 1080p (Q3) · up to 4K (Q4)
- Aspect ratios
- 9:16 · 3:4 · 1:1 · 4:3 · 16:9
- Sound
- Native dialogue, effects and music
- Inputs
- Text or a start frame (Q3) · a photo or reference images (Q4)
Made with Vidu on Affogato
Every clip below was generated with the prompt shown, unedited. Load any prompt into the generator and make your own version.
Anime
Anime skater at dusk
Prompt
Anime style, hand-drawn look: a girl with a short green bob and an oversized hoodie skateboards down a steep Tokyo street at dusk, weaving past a delivery bike. She ollies over a curb, lands, and grins back at the camera as cherry blossom petals swirl. Cel shading, warm orange sky, dynamic tracking camera. Sound: skateboard wheels, city ambience, an upbeat lo-fi track.
Reference to video
The gym still, brought to life
Prompt
The man from [@reference_image_1] does a slow dumbbell curl in the same gym, sets the weight down, picks up the black shaker bottle and raises it to the camera with a grin, saying "day one."
Image to video
The Paris photo keeps walking
Prompt
She keeps walking across the boulevard, the trench coat swinging with each step, then turns her head toward the camera and smiles as a bus passes behind her. Natural street sound.
UGC unboxing
Sneaker unboxing, one line
Prompt
Vertical UGC video: a young woman with pink braids sits cross-legged on her bed, lifts the lid off a sneaker box and pulls out a pair of chunky white sneakers. She holds one up to the camera and says, "okay, these are even better in person." Natural window light, slight handheld movement. Sound: tissue paper rustle, her voice.
Short drama
A forgotten anniversary, two shots
Prompt
A short vertical drama in two shots. Shot one: in a small kitchen at night, a woman in her thirties sets down a mug a little too hard and says, "you forgot again." Shot two, close-up on her partner, a man with tired eyes, who smiles sheepishly, pulls a small wrapped gift from behind his back and says, "did I?" She bursts out laughing. Warm pendant light. Sound: their voices, a fridge hum, no music.
Couples dance
A couple's spin on the dance floor
Prompt
Vertical shot of a couple dancing salsa in a crowded dance hall: he spins her twice, she dips back and laughs, her red dress flaring. Warm stage lights, other dancers blurred around them. The camera circles slowly. Sound: a lively salsa band, shoes on the wooden floor.
Cuts in one clip
Rainy cafe in three shots
Prompt
A cafe scene in three shots: a wide shot of a busy corner cafe in the rain; a medium shot of a young man at the window table sketching in a notebook; a close-up as a woman sits down across from him, he looks up and smiles. Soft daylight, steam from cups. Sound: rain, cafe chatter, a cup set down.
Which Vidu to pick
Start here
Vidu Q3
Text prompts, anime and drama
Text or a start frame in; a clip with sound and cuts out. Start here for anything you describe in words.
5 s clip 140 credits · 5–10 s
Open with Vidu Q3Option 2
Vidu Q4
Bringing your photos to life
Animate a photo or keep a character from reference images. Needs an image; there is no text-only mode.
5 s clip 87 credits · 4–16 s
Open with Vidu Q4Earlier Vidu versions are not offered on Affogato. Credits are read from the live catalog; the Standard plan ($24/month) includes 2,000 credits, about 14 five-second Vidu Q3 clips.
Prompting Vidu well
- Name the style first for stylised clips: anime, hand-drawn, cel shading.
- Describe the shot sequence when you want cuts, and keep the whole beat inside the clip length.
- Put dialogue in quotes and add the reaction you want as its own sentence.
- With Q4 references, tag them in the prompt ("the man from [@reference_image_1]") and say what he does.
- Name the camera move ("camera slowly orbits") and the light.
From prompt to clip
Describe the scene or add a photo
Write the scene in the box at the top of this page, or start from an example. Add a photo to use it as the first frame.
Sign in and land in Video Studio
Your prompt, photo and format carry over with Vidu selected. Nothing runs until you press Generate.
Pick the length, then generate
The credit price shows under the Generate button before anything is spent.
Switch to Q4 for your own characters
Use Vidu Q4's reference mode to keep a person, outfit or product from your images.
Questions about Vidu
What is Vidu best at?
Anime and stylised clips, short dramas with dialogue and sound, and (with Vidu Q4) animating your own photos and characters.
What is the difference between Vidu Q3 and Q4?
Q3 takes a text prompt or a start frame. Q4, a preview of Shengshu's next model, works only from a photo or reference images, and keeps people and products consistent.
Does Vidu generate sound?
Yes. Dialogue, effects and music come with the picture.
Who makes Vidu?
Shengshu Technology, a Beijing AI company.
Is Vidu free on Affogato?
No. Every run costs credits, priced by clip length and shown before you generate. Credits come with a monthly plan and unused credits roll over up to the plan's cap.
Can I use Vidu videos commercially?
Yes. Paid Affogato plans include a commercial license and no watermark, within our terms of service.
Keep exploring
Facts checked November 15, 2026. Sources: TechNode: Vidu Q3 · Vidu: Q4 · Unite.ai: Vidu Q4 Preview · WaveSpeed: Vidu Q3 review · fal: Vidu Q3 · fal: Vidu Q4 reference to video