Shengshu Technology Closes Over RMB 600 Million Series A+ Round: Reshaping AI Productivity Across Digital and Physical Worlds, Powered by Multimodal Foundation Models

ShengShu Technology was one of the earliest teams globally to conduct research on multimodal generative algorithms. In September 2022, the team introduced the U-ViT architecture. In July 2024, Vidu launched globally, introducing the industry’s first Reference-to-Video capability. Moving beyond conventional text-to-video and image-to-video models, this innovation addresses one of the core challenges in commercial video generation: maintaining multi-entity consistency. Since its launch, ShengShu has continuously released Vidu Q1, Vidu Q2, and Vidu Q3, with each iteration further improving performance across key benchmarks, including consistency, semantic understanding, motion dynamics, stability, and inference speed.

The recently released Vidu Q3 is the world’s first video model built for storytelling. It supports 16-second synchronized audio-video generation, native 1080p output, advanced cinematic language, precise shot transitions, multilingual text rendering, and multi-language output.

According to the latest rankings from AI benchmarking authority Artificial Analysis, Vidu Q3 ranked No.1 in China and No.2 globally. The model placed ahead of several leading international video generation platforms, positioning Vidu among the world’s top-tier solutions. Artificial Analysis data also indicates that Vidu Q2 maintains the fastest generation speed globally among commercial-grade content generation models.

In December 2025, ShengShu Technology open-sourced its TurboDiffusion framework, enabling a 5-second video to be generated in just 1.9 seconds on a single RTX 5090 GPU, improving video generation efficiency by 100 to 200 times.

ShengShu Technology has established a comprehensive product ecosystem built around Vidu, including Vidu MaaS, Vidu SaaS, Vidu App and Vidu Agent, serving content creators and industry clients globally. In 2025, the company achieved more than 10× growth in both users and revenue. Vidu is now widely used by creators, studios, and enterprises in over 200 countries and regions worldwide for content production.

In film and entertainment, including animation, short drama, and feature production, Vidu works with over 90% of industry stakeholders across content owners, tool providers, and production studios. Clients and partners include Tencent Animation & Comics, China Literature, CCTV Animation, iQIYI, Jiangxi Film Group, Mango TV, Soy Sauce Animation, and JuriLu.

Vidu continues to expand across leading companies in internet and smart hardware sectors. Customers include ByteDance, SAMSUNG, Wondershare, TAL Education Group, Alipay, and HONOR, with applications spanning content production, marketing asset creation, and interactive product innovation.

In advertising, Vidu has become a preferred solution for leading brands, agencies, and platform operators, working with JD.com, Alibaba 1688, Amazon, Meituan, Focus Media, BlueFocus, L’Oréal, and Anta. The platform supports large-scale and reliable AI video production for commercial campaigns.

In gaming, Vidu has been deployed in core scenarios such as game advertising and scene generation, serving companies including Lilith Games, 37 Interactive Entertainment, SeaArt, and Tanwan Games, enhancing content production efficiency and interactive experiences.

Internationally, Vidu continues to gain traction among global creators and enterprise clients, including Pollo AI, PhotoGrid, OpenArt, Hubx, Fal.ai, Eachlabs, Freepik and GensPark. Its applications span creative tools, commercial production, and platform-level services, with additional adoption across education, broadcasting, and cultural tourism sectors.

“ShengShu Technology began with a strong foundation in algorithm research and continues to push the boundaries of core model innovation. Among leading international multimodal foundation models, Vidu has established clear differentiation and strong competitive advantages. From research breakthroughs to large-scale commercialization, we believe multimodal foundation models will become a next-generation production paradigm and a transformative force in productivity, reshaping global content workflows and industry structures. ShengShu will remain technology-driven and value-oriented, advancing product and commercial strategies to unlock the full potential of multimodal AI for the global content ecosystem,” said Yihang Luo, CEO of ShengShu Technology.

“The ceiling for multimodal video models is exceptionally high. Beyond powering digital content creation and interaction, they have the potential to evolve into true world models that understand the underlying structures of reality and support end-to-end machine decision-making. As AI continues to mature in the digital world, ShengShu aims to push its boundaries further, expanding from digital deployment toward deeper integration with the physical world,” said Jun Zhu, Founder and Chief Scientist of ShengShu Technology.