Vidu
: How to use it, features, and the business problems it solves
Add bookmark
What is Vidu?
An AI video generator developed by ShengShu Technology in China. It creates videos from text, images, and multiple reference images, and its "Reference to Video" keeps characters and props looking consistent. Vidu Q3 generates video with audio up to 16 seconds, and Vidu Q4 Preview (October 2026) supports up to 4K and 15 reference images. The official site is available in Japanese; pricing is in USD from $10/month (free credits available).
Business problems it solves
About "Vidu"
What is Vidu?
Vidu is an AI video generator that creates videos from text, images, and multiple reference images. It is built on proprietary video models developed by ShengShu Technology in Beijing, China, and its standout feature is "Reference to Video": you upload several images of characters, props, and backgrounds, and Vidu animates them while keeping their appearance consistent. It is also well regarded for anime-style output, and the official site features testimonials from Japanese creators.
Models are named by generation with a "Q" number. Vidu Q3 (January 2026) added video with native audio—dialogue, sound effects, and music generated together with the picture, up to 16 seconds. On October 7, 2026, the next-generation flagship Vidu Q4 Preview was released, supporting up to 4K output and up to 15 reference images and 3 reference audio clips. The official site (vidu.com/ja) and pricing page are available in Japanese, but prices are in US dollars.
Three ways to create: text, image, and reference
Vidu offers three main generation modes. Which one you choose determines how much control you have over the result.
Text to Video
Describe the scene in words. You can start without any assets
Image to Video
Animate a single still image. You can set start and end frames to define the motion
Reference to Video
Provide several images of characters, props, and backgrounds, and keep the same look in every shot
Characters and props used in Reference to Video can be saved to "My References" and reused in other videos. This reduces the problem of a character's look changing from shot to shot in series-style short videos and anime.
How to use
-
Create an account
Sign up on the official site (vidu.com/ja). New users receive trial credits, and you also earn credits by logging in daily. The service is for users aged 18 and over.
-
Choose a generation mode
Pick "Text to Video," "Image to Video," or "Reference to Video." Use Reference to Video to lock in a character, or Image to Video to animate a single image you already have.
-
Add assets and a prompt
Upload images or reference assets, and describe the motion, camera work, and dialogue in the prompt. Select the model (Q4, Q3, etc.), length, and resolution.
-
Generate and review
Run the generation and check the result. If it isn't what you wanted, adjust the prompt or reference images and try again. If a Japanese prompt doesn't come across as intended, rewriting it in English is worth a try.
-
Download and use
Download the finished video. Paid plans can download without a watermark.
Features
01Generate
- Text to Video — Generate video from written instructions
- Image to Video — Turn a single image into a 3–16 second video (Q4). Start and end frames can also be specified
- Reference to Video — Reference 1–15 images and 0–3 audio clips to keep characters, style, scene, and voice consistent (Q4)
- Image generation — Create still images from text or reference images (Reference to Image / Text to Image)
02Direct
- Video with audio — Generate dialogue, narration, sound effects, and music together with the picture, with automatic lip sync (Q3)
- Multi-character dialogue — Create conversations between characters in a single generation (Q3)
- Camera control — Specify camera movement and pacing for each scene
- AI sound effects & templates — Generate sound effects, or use ready-made templates for popular effects
03Extend
- Off-Peak Mode — A generation allowance that runs during less busy hours: up to 20 videos a day on Premium and 200 a day on Ultimate
- Vidu Agent & app — An agent that works with you from a single line to a finished piece, plus a smartphone app
- API Platform — An API for developers and businesses (platform.vidu.com) to build Vidu into their own services
Pricing
Vidu is credit-based: each video generation consumes credits. Consumption depends on the model, resolution, and length. For example, Vidu Q4 (540p) costs 2 credits per second on paid plans but 4 credits per second on the free plan, so the free plan is effectively more expensive per second.
| Plan | Monthly billing | Annual billing (per year) | Monthly credits | Key differences |
|---|---|---|---|---|
| Free | $0 | — | Granted at sign-up and for daily logins | Non-commercial. 10 My References per month |
| Standard | $10 | $99 ($8.25/month) | 800 | Commercial use, no watermark, 4 concurrent generations, 300 My References per month |
| Premium | $43 | $425 ($35.42/month) | 3,500 | Standard plus the fastest channel and 20 off-peak videos/day |
| Ultimate | $115 / $230 | $1,150 / $2,300 | 10,000 / 20,000 | Premium plus 200 off-peak videos/day. Choose between two credit tiers |
Refunds are not offered. The FAQ on the official pricing page states that refunds are not accepted. Subscriptions made through Stripe renew automatically every month or year; to stop automatic payment, contact support (support@vidu.com). Subscription credits included in a plan are valid for 30 days, while purchased credits and bonus credits from events are valid for 2 years.
Promotional pricing appears frequently. When checked in October 2026, annual billing showed a "first-time only, up to 17–18% off" label, and the homepage announced discounts for the Q4 launch (up to 47% off annual plans and up to 40% off monthly plans). The table above shows the regular, undiscounted prices.
Prices and credits are based on the official pricing page (individual plans) as of October 14, 2026. Prices are in US dollars and may change with exchange rates, promotions, and plan revisions, so please check the official pricing page for the latest information.
Reviews and reputation
- In third-party evaluations, Vidu Q4 Preview ranks 2nd on Artificial Analysis's image-to-video leaderboard (with audio), which ranks video models by human preference votes (as of October 14, 2026; the top five models are within the margin of error).
- Testimonials on the official site praise "highly expressive animation that conveys subtle expressions and nuance," recommend it "if you want to make anime with conversation scenes," and call it "great value because you can try Q3 with audio for just a few credits."
- On the other hand, the free plan consumes credits at twice the rate of paid plans, so serious production effectively requires a paid plan. The no-refund policy is also worth noting before you subscribe.
Things to check before adopting
Vidu is developed by a Chinese company and operated by a Hong Kong entity. If you plan to use it for work, compare the following points with your organization's standards.
01Where data is stored
- Stored in Singapore, processed in regions including China — The privacy policy states that user information is stored in Singapore and may be processed in other jurisdictions, including China
- Governed by Hong Kong law — The terms of service are governed by Hong Kong law, with disputes resolved by arbitration at the Hong Kong International Arbitration Centre (HKIAC)
02How uploaded assets are handled
- The official FAQ and the terms differ — The FAQ on the homepage says uploaded content "is not used for AI training," but the terms of service (updated July 3, 2026) include a clause granting Vidu a license to use user-submitted content for product improvement, including AI training, and for promotional purposes
- Don't upload confidential material — Before using unreleased product images or confidential assets, it is safest to review the latest terms and decide according to your internal policies
03Rights to outputs
- Commercial use on paid plans only — Under the terms, the free plan is for personal, non-commercial use, and paid plans permit commercial use
- You own what you create — Under the terms, Vidu does not claim ownership of content you create. However, depicting real people or existing characters requires separate checks on portrait rights and copyright
How Vidu differs from other AI video generators
| Aspect | Vidu | Other major models (Kling / Hailuo / Sora / Veo, etc.) |
|---|---|---|
| Strengths | Character and prop consistency via reference images; anime-style output | Varies by model |
| Reference assets | Up to 15 images + up to 3 audio clips (Q4) | Varies by product |
| Audio | Dialogue, sound effects, and music generated together (Q3 and later) | Varies by product |
| Pricing | From $10/month (USD, free credits, no refunds) | Varies by service |
Compared with fellow Chinese services Kling, Hailuo, and Seedance, or with Sora, Veo, and Runway ML, Vidu tends to be chosen when you want to reuse the same characters across many videos. It is a strong option for anime, short dramas, and series-style ads, where visual consistency makes or breaks the result.
Frequently asked questions (FAQ)
Q. Can I use Vidu for free?
Yes. New users receive trial credits, and you also earn credits by logging in daily. However, the free plan consumes more credits than paid plans (4 credits per second for Q4 at 540p), and under the terms it is limited to personal, non-commercial use.
Q. Can I use it in Japanese?
The official site and pricing page are available in Japanese (vidu.com/ja). Vidu Q3 supports generating videos with Japanese dialogue. You can write prompts in Japanese, but if the result isn't what you intended, rewriting the prompt in English can make it easier to fine-tune.
Q. How much does it cost? Can I pay in Japanese yen?
Paid plans are Standard at $10/month, Premium at $43/month, and Ultimate at $115 or $230/month (as of October 2026), billed in US dollars. Prices are not shown in yen; charges are converted at your credit card's exchange rate.
Q. Can I cancel or get a refund?
The official pricing page states that refunds are not offered. To stop automatic payment, contact support (support@vidu.com). Because annual billing involves a large upfront amount, it is safest to start with a monthly plan.
Q. Can I use the videos commercially?
Videos generated on paid plans can be used commercially and downloaded without a watermark. The free plan is limited to personal, non-commercial use. When depicting real people or existing characters, you need to check portrait rights and copyright separately from Vidu's terms.
Q. What is the difference between Vidu Q3 and Vidu Q4?
Vidu Q3 (January 2026) is the model that introduced video with native audio—dialogue, sound effects, and music generated together with the picture, up to 16 seconds. Vidu Q4 Preview (October 2026) is the next-generation flagship, supporting up to 4K output, Reference to Video with up to 15 reference images and 3 reference audio clips, and 3–16 second Image to Video.
Q. What is the difference between Reference to Video and Image to Video?
Image to Video animates a single image. Reference to Video takes multiple images of characters, props, backgrounds, and so on, and creates new scenes while keeping their appearance consistent—ideal when you want to use the same character across multiple videos.
Q. Is it safe to use?
The operator is ShengShu AI HK Limited in Hong Kong, and the privacy policy states that user information is stored in Singapore and may be processed in regions including China. For business use, review the latest terms of service and privacy policy before uploading highly confidential material.
The information on this page is based on what was published on the official site (vidu.com), terms of service, and privacy policy as of October 14, 2026. Models, features, and pricing may change, so please check the official site for the latest information.
