Vidu Launches Q2 "Reference-to-Video", Pioneering a New Era of High Consistency and Creative Control

With better consistency, faster generation, and more affordable pricing, ShengShu Technology offers a top-tier alternative to leading global platforms.

SINGAPORE, Oct. 21, 2025 /PRNewswire/ -- ShengShu Technology, a global leader in multimodal generative AI, today launched Vidu Q2 "Reference-to-Video", setting a new standard for AIGC Products. This release shifts AI from simply creating motion to performing it, blending technology with creativity to deliver realistic, emotional, and cinematic videos.

The new "Reference-to-Video" feature allows creators to make consistent, expressive videos using up to seven reference images for faces, gestures, scenes, or props. It can combine multiple unrelated elements, like different characters, items, or backgrounds, into a single, unified video. With a text prompt, it blends these elements using its Multiple-Entity Consistency feature, keeping each part distinct and true to its original appearance, even in complex or changing scenes. With faster generation and more affordable pricing, Vidu Q2 "Reference-to-Video" now competes with the best video platforms, bringing the "AI performance era" closer.

"Vidu Q2 'Reference-to-Video' marks a new chapter in AI video creation," said Yihang Luo, CEO of ShengShu Technology. "We're moving into a time where AI can mimic human looks and express emotions with cinematic flair. This launch goes beyond basic video creation; it's about teaching AI to act and tell stories alongside creators."

Bringing Realism and Cinematic Quality to AI Video

Vidu Q2 "Reference-to-Video" delivers a significant leap in realism, capturing subtle emotions like a hesitant smile, a curious look, or tense anticipation with natural flow. Movements feel smooth and alive, replacing stiff, robotic motions with vibrant energy.

The platform also handles cinematic techniques like smooth camera shifts, panning, and depth of field, mimicking professional filmmaking. Scenes transition seamlessly between wide shots and close-ups, letting creators craft stories with both grandeur and emotional depth.

Vidu Q2 "Reference-to-Video" also better understands prompts, capturing the intended mood and meaning more accurately. This makes it easier for creators to focus on storytelling rather than fixing errors, turning generative video into a practical, everyday tool.

With faster generation and higher accuracy, creators can produce longer, expressive videos for film, animation, advertising, and commercial content with minimal adjustments.

Swift Market Adoption Proves Real-World Value

With the launch of Vidu Q2 "Reference-to-Video", the Vidu Q2 MaaS API was also made available globally, enabling businesses to easily add Reference-to-Video features to their work processes. Building on its established expertise in Reference Generation technology, ShengShu quickly formed a number of new partnerships with advertising and e-commerce companies, demonstrating the model's high-quality output and ease of use. The update helps businesses by providing new ways to reduce costs, improve efficiency, and enhance creative quality.

The model's high consistency is especially valuable for commercial video production. Even with complex camera movements or character interactions, the main subject and product details remain clear and stable, creating lifelike 360-degree displays that look just like real footage. In product advertising, models now display natural gestures and micro-expressions, producing highly realistic and engaging footage that redefines what AI-generated content can achieve.

A Continuing Story of Innovation

Since its founding, ShengShu Technology has led the way in creative AI with groundbreaking advances. In 2022, the team introduced the U-ViT architecture, the world's first Diffusion–Transformer hybrid model, before the DiT architecture used by top competitors. The company also created UniDiffuser, the first model to generate both text and images in one system. Building on these, ShengShu's Analytic-DPM framework made AI processing up to 20 times faster, earning the ICLR 2022 Outstanding Paper Award. Its DPM-Solver became the world's fastest diffusion solver, later used in platforms like Stable Diffusion.

Each step in the Vidu series turned these breakthroughs into creative platforms: Vidu 1.5 brought the world's first consistent multi-character scenes, Vidu 2.0 offered 10-second videos at half the industry cost, and Vidu Q1 added cinematic transitions with realistic sound. Now, Vidu Q2 "Reference-to-Video" combines these advances into one powerful, expressive system, moving the platform from AI creation to true AI performance.

A Rapid Rise to Global Top Player

Since Vidu's launch in April 2024, ShengShu Technology has seen rapid growth. In just one year, the platform reached over 200 countries, gained 30 million users, and produced over 400 million videos, cementing its place as a top player in creative AI. "With each release, we blend technology and creativity more closely," said Luo. "Our goal isn't to replace creativity but to expand it, making imagination visible and emotions limitless."

To learn more about Vidu, visit www.vidu.com

Vidu API is available at platform.vidu.com.

About ShengShu Technology

Founded in March 2023, ShengShu Technology is a world-leading artificial intelligence company, specializing in the development of Multimodal Large Language Models. Driven by innovation, the company delivers cutting-edge MaaS and SaaS products that revolutionize creative production by enabling smarter, faster, and scalable content creation. With its flagship video generation platform Vidu, ShengShu Technology's solutions have reached more than 200 countries and regions around the world, spanning fields including interactive entertainment, advertising, film, animation, cultural tourism, and more.

SOURCE ShengShu Technology