# Vidu

> Generate consistent AI videos from text, images and reference characters in one workspace

- Category: [Video AI](https://finderatlas.com/categories/video-ai)
- Pricing: Freemium
- Website: https://www.vidu.com
- Platforms: Web, iOS, Android, CLI, API
- API: available
- Mobile app: available
- Made by: ShengShu Technology, founded 2023, Beijing, China
- Last updated: 2026-10-19
- FinderAtlas page: https://finderatlas.com/tools/vidu

## Overview

Vidu is an AI video generator from ShengShu Technology that turns text prompts, still images and reference material into short video clips. The service supports Text to Video, Image to Video (including first and last frame control) and Reference to Video, along with Text to Image and Reference to Image. Its current flagship model is Vidu Q4, alongside Vidu Q3 and Vidu S, a real-time model that the company says responds in milliseconds.

## Key features

- Reference to Video
- My References
- Image to Video
- Text to Video
- Text and Reference to Image
- Vidu S real-time model
- Vidu Agent and AI Templates
- Off-Peak Mode
- API Platform

## Pricing

- **Free** ($0): Free credits for trying the generator
- **Standard** ($9.99/month): Entry paid tier for regular creators
- **Premium** ($19.99/month): Higher volume and quality for frequent creators
- **Ultimate** ($59.99/month): Largest credit pool for heavy users
- **Enterprise** (Custom): Teams and organisations

## Best for

- Short clips that need a consistent character or product across shots, and animating still images

## Who uses it

- Content Creators
- Marketers
- Filmmakers
- Social Media Managers
- Developers

## Strengths

- Reference to Video with a saved My References library helps keep characters and props consistent.
- Supports text, image and reference inputs, plus first and last frame control.
- Free credits let users test the product before paying.
- Public API with published per-second pricing for each model and resolution.
- Available on web, iOS, Android and a CLI.

## Limitations

- Plan prices are not visible without loading the pricing page, and third-party sources disagree on them.
- Credit costs vary by model and resolution, so per-video cost is hard to predict; 4K Q4 preview output costs 78 credits per second in the API.
- CostBench reports user complaints about credits being spent on failed renders in the consumer product, and a 2.2/5 Trustpilot score from 17 reviews.
- Clip generation only; scripting, sequencing and final mixing need other tools.

## Not recommended for

- Long-form video editing, or users who need fully predictable per-video costs

## FinderAtlas editorial scores (out of 10)

- Ease of Use: 8.2/10
- Feature Depth: 8/10
- Value for Money: 7.3/10
- Performance & Speed: 8/10
- Documentation & Support: 7.4/10

## Alternatives

- [PixVerse](https://finderatlas.com/tools/pixverse): Generate videos from text, images and prompts with AI models, agents and APIs ([compare](https://finderatlas.com/compare/pixverse-vs-vidu))
- [Pika](https://finderatlas.com/tools/pika): Generate, edit and extend AI video, images and audio from one creative platform ([compare](https://finderatlas.com/compare/pika-vs-vidu))
- [Higgsfield AI](https://finderatlas.com/tools/higgsfield): Generate AI video and images from 30+ models in one creative platform ([compare](https://finderatlas.com/compare/higgsfield-vs-vidu))
- [HeyGen](https://finderatlas.com/tools/heygen): Create avatar videos from text and translate them into 175+ languages ([compare](https://finderatlas.com/compare/heygen-vs-vidu))
- [Luma Dream Machine](https://finderatlas.com/tools/luma-dream-machine): Generate and edit video and images from text or images with Luma's Ray models ([compare](https://finderatlas.com/compare/luma-dream-machine-vs-vidu))
- [VEED](https://finderatlas.com/tools/veed): Browser-based AI video editor for subtitles, avatars, dubbing and social-ready videos ([compare](https://finderatlas.com/compare/veed-vs-vidu))
- [Captions](https://finderatlas.com/tools/captions): AI video editor that turns raw footage and prompts into finished, captioned videos ([compare](https://finderatlas.com/compare/captions-vs-vidu))

## Recent updates

- 2026-04-01: Press reports say ShengShu raised RMB 2 billion in a round led by Alibaba Cloud
- 2024-04-01: Vidu was first unveiled at the Zhongguancun Forum in Beijing

## FAQ

**Is Vidu free to use?**

Vidu gives new users a set amount of free credits, and paid subscriptions are available for more credits and watermark-free output. The exact free credit amount is shown inside the product.

**Who makes Vidu?**

Vidu is developed by ShengShu Technology, a Beijing-based AI company founded in 2023, with a research link to Tsinghua University. The official mobile app is published under ShengShu AI HK Limited.

**Can I keep the same character across several clips?**

Yes. Reference to Video accepts multiple images of a subject, and My References lets you save characters, props and scenes for reuse.

**Does Vidu have an API?**

Yes. Enterprises and developers can use the API platform at platform.vidu.com, which is billed in credits that vary by model and resolution.

**Are there official mobile apps?**

Yes. The Vidu - AI Video Generator app is on Google Play (package com.vidu.oversea) and the App Store. Several unrelated apps use similar names, so check that the publisher is ShengShu.