
Shengshu Releases Vidu 2.0
Shengshu Technology recently released its version of a high-end video generation platform: Vidu 2.0. The newly developed technology, with its current upgrade, looks forward to completely revolutionizing the future of content creation using artificial intelligence. As stated by the company, with Vidu 2.0, the world will experience quicker, cheaper, and more convenient video production.
The feature of Vidu 2.0 stands out due to its capability of producing video clips in just 10 seconds, an unprecedented leap in efficiency. Moreover, the cost-effectiveness of producing videos on this platform is more than 55% less than that of the industrial standards, therefore wider ranges of people can access it. While the average industry cost lies at about USD 0.084 per second of video, Vidu 2.0 returns the same for about USD 0.0375 per second, therefore cutting down significantly and outdoing its closest predecessor, Vidu 1.5.
The heart of Vidu 2.0 is the proprietary Universal Vision Transformer (U-ViT) model developed by Shengshu, along with a full-stack interference accelerator. This combination is powerful enough to allow the platform to deliver high-quality video outputs at unprecedented speeds and costs, setting a new benchmark in AI video generation.
Simplified Video Creation for All Users
Vidu 2.0 is an intuitive feature that simplifies the video-creation process with pre-configured templates, which are also referred to as preset prompts. Such templates enable users to add easily detailed actions, props, or settings to their videos without writing complex prompts by hand. For example, one can choose outfits or environments from a comprehensive library, making it user-friendly for even beginners. Shengshu has also announced plans to expand this library, giving creators more options over time.

This latest update is a reflection of Shengshu’s broader mission to pioneer AI-driven storytelling that dynamically adapts to the needs and creativity of its users. The platform’s advancements align with the growing global interest in generative AI, positioning Shengshu as a leader in the field.
Shengshu’s Journey and Competitive Landscape
Founded in March 2023, Shengshu focuses on developing deep generative models by intertwining diffusion techniques with transformers to specialize in multimodal applications. It has transitioned from research to commercialization by launching Vidu in July 2024. Apart from Vidu, Shengshu also developed VoxCraft, an asset generation platform for 3D.
Vidu 2.0’s launch is timely with the increase in generative AI innovations all over the world. OpenAI has recently opened up public access to its video generation tool, Sora. Google has included its Veo 2 model into VideoFX and opened it to selected users in a waitlist. In China, Shengshu competes with the homegrown giants, including Kling AI from Kuaishou, Hailuo AI of MiniMax, and Dreamina of ByteDance.