Text to video, 1 to 16 seconds
Vidu Q3This page
Write a prompt and get a clip, no image needed. Any whole second from 1 to 16 at 540p, 720p or 1080p, with Pro for the best results and Turbo for speed and price. Turbo also animates a still.
AI Video Generator
Vidu Q3
Write the scene, the lines and the sound, and Shengshu's Q3 returns all three together, from one second to sixteen.
Output
Sample
Vidu Q3 or Vidu Q4 Preview?
Both come from Shengshu and both generate sound with the picture. Q4 Preview is the newer model and ranks well above Q3 Pro for image to video, but it only starts from pictures. Q3 is the one that works from text alone and goes down to one second. Five seconds at 720p cost 113 credits on Q3 Turbo, 193 on Q3 Pro and 159 on Q4 Preview.
Image to video and reference to video
Vidu Q4 Preview
Start from a first frame, or from up to 15 reference images and 3 voice clips that keep a cast and its voices together. Clips run 3 to 16 seconds, from 540p up to 2K and 4K. No text to video yet.
Open the Vidu Q4 Preview pageWhat Vidu Q3 does well
01
Two lines, two voices
Two friends on a rooftop at sunset. She says she is moving to Tokyo next month, he answers that this is their last summer, both in Japanese. Each line comes out in its own voice, on the face of the person saying it, with a cut between two close-ups because the prompt gave each speaker a shot.
Settings
- Text to video
- Pro
- 8s
- 720p
- 16:9
- Audio on
2D anime, hand-drawn cel shading. Sunset on the flat roof of a small seaside apartment building, laundry lines and a water tank, orange sky over the sea. Two friends in their twenties: a woman with a short black bob in a white T-shirt and a man with messy brown hair in a grey hoodie. Shot 1, 0-4s: close-up on the woman, she looks out at the sea and says in Japanese: 「来月、東京に引っ越すんだ。」 Shot 2, 4-8s: hard cut to a close-up on the man, he looks at her, surprised, then smiles and answers in Japanese: 「じゃあ、最後の夏だね。」 Only the person on screen speaks. Sound: their two voices, gulls, distant waves, a soft wind chime, no music.
02
Two seconds, one pop
A cork leaves a champagne bottle and the pop lands on the frame it should, with the foam and the fizz after it. Q3 makes clips as short as one second, the length of an insert or a transition. Vidu Q4 starts at three.
Settings
- Text to video
- Turbo
- 2s
- 720p
- 16:9
- Audio on
Extreme close-up on a black background: the cork shoots out of a chilled green champagne bottle and white foam bursts over the neck, droplets catching a rim light. Locked-off camera, high-speed commercial look. Sound: one sharp pop and a fizz.
03
A paper boat in one take
A paper boat drifts down a forest stream, knocks a mossy stone, turns and carries on into the light, ten seconds without a cut written as a single paragraph. The piano, the dripping leaves and a distant bird come from the same generation.
Settings
- Text to video
- Turbo
- 10s
- 720p
- 16:9
- Audio on
A folded white paper boat drifts down a narrow forest stream after rain. The camera tracks low beside it as it bumps a mossy stone, spins once and floats on under drooping ferns into a patch of sunlight. One continuous shot, no cuts, photorealistic, shallow depth of field. Sound: trickling water, drops falling from leaves, a distant bird, soft solo piano.
What you can make with Vidu Q3
Scenes for a comic drama
Give each speaker a shot and a line, in English, Japanese or Chinese, and cut the result straight into an episode.
Inserts and transitions
One to three second shots for the gaps in an edit: a door shutting, a glass set down, a cork popping, each with its own sound.
A clip from a sentence
No image to start from: write the scene and the sound, pick a length up to 16 seconds, and picture and sound come back together.
The technology behind Vidu Q3
How the model works, based on public documentation and sourced evidence.
Vidu Q3 Pro comes from Shengshu Technology, a company spun out of Tsinghua University research. The Vidu model family is a diffusion model built on U-ViT, a transformer backbone for diffusion that Shengshu's researchers proposed in 2022, before the DiT design later popularized by Sora. According to Shengshu's technical report, this architecture is what lets Vidu scale to longer, more consistent video.
Launched in early 2026, Q3 was Shengshu's first model to generate sound and picture together in one pass instead of adding audio afterward. Shengshu pitches it on dialogue, voiceover, sound effects and music that come out with the video, on conversations between several speakers, and on output in English, Japanese or Chinese, with up to 16 seconds in a single run. The Pro tier adds what Shengshu calls storyboarding: one prompt can lay out several shots, and the model cuts between them inside the same clip.
On Vidney, Q3 comes in two tiers. Pro runs text to video. Turbo, which Shengshu describes as the faster of the two, runs text to video and image to video. Both take any whole number of seconds from 1 to 16 at 540p, 720p or 1080p, and text to video offers 16:9, 9:16, 4:3, 3:4 and 1:1. Sound is off until you switch it on, and switching it on does not change the price. Start and end frames and the reference mode Shengshu added in April 2026 are not offered here.
Vidu Q4 Preview arrived on October 7, 2026 and debuted sixteen places above Q3 Pro on Artificial Analysis' image to video leaderboard. Q3 keeps two things Q4 does not do: text to video, where a prompt alone makes the clip, and clips shorter than 3 seconds. Turbo is also the cheaper of the two at 720p. For a clip that starts from a still, a cast held together by reference images, or 2K and 4K output, Q4 is the newer choice.
Sources: prnewswire.com · arxiv.org · artificialanalysis.ai
Who reaches for Vidu Q3
Vidu Q3 is Shengshu's sound and picture model from early 2026, built, in Shengshu's words, for comic dramas, films and short series. It works from a prompt alone: describe the scene, the lines and the sound, and dialogue, effects and music come back with the picture, in English, Japanese or Chinese. Clips run any whole number of seconds from 1 to 16 at 540p, 720p or 1080p. Pro is the tier Shengshu tunes for results; Turbo is faster, costs less and can also animate a still. People who keep Q3 open next to Vidu Q4 Preview want one of two things the newer model does not do: a clip from text with no picture to start from, or a shot shorter than 3 seconds for an insert, a transition or a reaction. Sound is off until you switch it on, and switching it on does not change the price.
How to generate video with Vidu Q3
Create AI video with Vidu Q3 on Vidney in three steps.
Vidu Q3 vs Vidu Q4 Preview vs Grok Imagine Video 1.5 vs Grok Imagine Video 1.5 Lite
Specs side by side with similar models, all runnable on Vidney.
| Vidu Q3 | Vidu Q4 Preview | Grok Imagine Video 1.5 | Grok Imagine Video 1.5 Lite | |
|---|---|---|---|---|
| Task modes | Text to Video · Image to Video | Image to Video · Reference to Video | Text to Video · Image to Video · Reference to Video | Text to Video · Image to Video |
| Duration | 1–16 seconds | 3–16 seconds | 1–15 seconds | 1–15 seconds |
| Resolution | 540p · 720p · 1080p | 540p · 720p · 1080p · 2K · 4K | 480p · 720p · 1080p | 480p · 720p · 1080p (upscaled from 720p) |
| Audio | Supported | Supported | Included, always on | Included, always on |
| Cost | 75–90 credits per generation | 159 credits per generation | 18 credits per generation | 18 credits per generation |
What creators say about Vidu Q3
Recurring themes from public reviews, comparisons and creator write-ups. Both the good and the bad.
What gets praised
Creators consistently praise the native audio generation: dialogue, sound effects and music come out already synced with the picture, and reviewers say it is the first time they did not have to fix audio timing in post.
Anime output is a recurring standout in reviews, with strong Japanese lip-sync and noticeably better character consistency in anime style than in realistic style.
The 16 second length combined with Smart Cuts gets frequent praise for making clips feel professionally edited, with the model cutting between wide shots and close-ups on its own instead of holding one static shot.
Common complaints
A common complaint is character consistency breaking down in busy scenes: with more than two characters on screen, faces and features can shift or blend between them, especially in realistic style.
Users also report prompt details getting dropped and unprompted camera moves creeping in, which means re-rolling a clip to get the shot that was asked for.
Related models
More AI video models you can run on Vidney.
Live
Shengshu's next flagship, out early: a still or a stack of references in, a scene with its own sound out.
159 credits per generation
Generate AI video with Runway Gen-4 Turbo: text to video and image to video, all from one Vidney workspace.
21 credits per generation
Live
Generate AI video with MiniMax H3 Max: text to video and image to video, all from one Vidney workspace.
48 credits per generation
Live
Generate AI video with Seedance 1.5 Pro: text to video and image to video, all from one Vidney workspace.
10–19 credits per generation
Live
Generate AI video with Veo 3.1: text to video and image to video, all from one Vidney workspace.
24–384 credits per generation
Every clip comes back with its own sound, dialogue and lip sync included.
18 credits per generation
Frequently asked questions
What is Vidu Q3?
Vidu Q3 is an AI video model you can run on Vidney. It supports text to video and image to video from a single workspace.
How much does Vidu Q3 cost per generation?
Vidu Q3 costs 75–90 credits per generation. The exact credit cost is shown on the generate button before you run it. That is about $2.25–$2.70 per run at the Starter plan rate ($0.03 per credit).
What aspect ratios does Vidu Q3 support?
Vidu Q3 supports the following aspect ratios: 1:1, 4:3, 3:4, 16:9, 9:16.
How long can Vidu Q3 videos be?
Vidu Q3 generates clips of 1–16 seconds.
Does Vidu Q3 generate audio?
Yes. Vidu Q3 can generate video with an audio track, and you can toggle sound on or off before generating.
Is Vidu Q3 free to use?
Not with sign-up credits alone. New accounts get 10 to 30 free credits, and Vidu Q3 costs 75–90 credits per generation, so running it needs a plan. You can spend the free credits on lower-cost models first; the exact cost is always shown on the generate button before you spend a credit.
Can I use Vidu Q3 videos commercially?
Yes. The videos you generate with Vidu Q3 on Vidney are yours to use in commercial projects: ads, social posts, client work, and product content. You keep the output; Vidney only handles generation and storage.
Does Vidu Q3 add a watermark?
Only on the free plan. Free-plan downloads carry a small Vidney watermark; on any paid plan, Vidu Q3 videos download clean, ready to publish or hand to a client. Upgrading also removes the watermark from videos you already made.
What are the best Vidu Q3 alternatives?
The closest Vidu Q3 alternatives on Vidney are Vidu Q4 Preview, Grok Imagine Video 1.5, and Grok Imagine Video 1.5 Lite. They all run in the same workspace from one credit balance, so you can run the same prompt on each and compare the results side by side.
Do I need an API key or a provider account to use Vidu Q3?
No. Vidney runs Vidu Q3 for you, so there are no API keys to manage and no separate provider account to set up.
