Singapore, January 30, 2026 /PRNewswire/ — ShengShu Technology, a global leader in multimodal generative AI, announced a series of significant upgrades to its AI video platform Vidu over the course of a week. Global Creativity Weekhighlighting efforts to move AI video from experimentation to large-scale production use.
These releases are led by Vidu Q3, the industry’s first long-form AI video model that enables native audio and video generation in a single output. Designed for narrative production, Q3 integrates sound and visuals directly at the model level, marking the transition from silent visual generation to fully synchronized storytelling. In the latest rankings published by AI benchmarking authority Artificial Analysis, Vidu Q3 was ranked #1. China And number 2 in the world. This model ranks Vidu above several major international video generation platforms and positions Vidu as one of the world’s top solutions.
Alongside Q3, the company announced the Vidu Q2 Reference-to-Video Pro, the world’s first video model where everything acts as a reference. Q2 Pro extends AI video from one-shot generation to structured, reference-driven workflows for greater control, consistency, and production efficiency. Upgraded with Reference Hub 2.0 vidu agent Additionally, we extend the platform into an integrated production system designed for professional creators and corporate teams.
As the global debate continues over whether breakthroughs in AI are leading to sustainable productivity, Vidu’s strategy is tochina speed” — quickly transform cutting-edge research into deployable systems embedded directly in production workflows.
“Advanced AI only makes sense if it creates tangible productivity,” he said. Luo YihangCEO of ShengShu Technology. “Our focus is not just on the functionality of the model, but on how quickly that functionality can be used in daily production.”
from AI function AI productivity: An unprecedented production breakthrough
During Global Creativity Week, Vidu announced a suite of tightly integrated upgrades aimed at transforming AI video from an experimental feature to a production-ready infrastructure.
Vidu Q2 Reference from Video Pro This represents a decisive step towards controllable creation. The world’s first video model that lets you use anything as a reference, supporting 2 video references and 4 image references in a unified workflow. Creators can combine inputs for people, scenes, actions, expressions, effects, and textures to add, remove, or change elements without having to regenerate the entire sequence. By moving from one-shot generation to reference-driven editing, Q2 Pro turns AI video into an iterative collaborative tool suitable for professional production environments where revision speed and consistency are key.
Buildings for storytelling, Vidu Q3 Proceed with synchronized audio-video generation directly at the model level. This is the industry’s first long AI video generation model. 16 seconds of native audio video Output in a single pass. Sound and vision are generated directly from the model, providing high-resolution, synchronized output. Q3 supports multilingual audio generation, precise lip sync, cinematic camera control, seamless shot transitions, and native 1080p rendering, with text generated as part of the visual composition rather than added in post-production. These features position Q3 for animation, short drama, film production, and other story-driven formats.
Vidu also introduced Reference Hub, the world’s first AI video subject-based creation community. Built on Vidu’s reference-driven video generation framework, creators can choose from eight structured subject categories to create more consistent, efficient, and professional-looking videos. At launch, over 200 carefully selected themes will be available, including camera movement, composition, narrative structure, visual style, scene design, character performance, abilities, and atmosphere. By standardizing and modularizing visual elements, the platform lowers the barrier to high-quality output, enables scalable reuse of creative assets, and introduces a new paradigm for AI-powered video creation.
To further streamline your production workflow, vidu agent 1.0 One-click video creation with intelligent planning and real-time generation. The upgraded agent supports custom background music upload, voiceover removal, and storyboard-based editing, reducing coordination complexity and accelerating end-to-end production.
Depth of research, speed of engineering
Vidu’s fast product pace reflects ShengShu Technology’s engineering-driven approach. Its early U-ViT architecture pioneered the integration of diffusion and transformer frameworks and laid the early foundations for modern video generation models. In late 2025, the company jointly released TurboDiffusion with Tsinghua University’s TSAIL Lab to significantly speed up inference by up to 200x while maintaining production quality.
Over the past year, Vidu has rolled out a series of model upgrades to achieve benchmark-leading performance in text-to-video generation and expand the creation of a highly consistent reference base. These advances culminate in Q2 Pro’s reference-driven precision and Q3’s native audio/video integration.
Expanding global adoption
Since its release April 2024Vidu has expanded to users in more than 200 countries and territories, serving more than 40 million creators and 10,000 developers and enterprise customers. The platform has generated over 500 million videos, with commercial projects accounting for over 70% of the total output.
Across the industry, creative platforms like PhotoGrid are incorporating Vidu functionality into their core products. Polo AI, an AI workflow platform, applied Vidu’s image-to-audio-video system to address long-standing synchronization challenges. Production teams used Vidu on long-form narrative projects, and fashion brand Odin deployed the technology to power its AI-driven virtual try-on experience.
“AI only makes sense when it’s used at scale in real-world workflows,” Luo says. “Global Creativity Week is another step in turning AI video innovation into practical creative infrastructure.”
