TLDR: Singapore-based AI video startup, Video Rebirth, has successfully closed a $50 million seed funding round. Led by former Tencent Distinguished Scientist Dr. Wei Liu, the company aims to develop a “world model” for AI video generation, focusing on high-fidelity, controllable platforms for professional creators in film, advertising, and animation. The funding will accelerate research into its proprietary “Physics Native Attention” architecture, with a 1.0 product launch anticipated in December 2025 to challenge existing consumer-grade tools and competitors like OpenAI’s Sora.
Video Rebirth, a pioneering AI video startup headquartered in Singapore, has announced the successful completion of a substantial $50 million seed funding round. This significant capital injection is earmarked to propel the company’s ambitious vision of building a “world model” for advanced AI video generation, specifically targeting the demanding needs of professional creators in sectors such as advertising, e-commerce, film, and animation. The funding round saw participation from a consortium of top-tier global financial and strategic investors, including notable contributions from Qiming Venture Partners and South Korean game developer Actoz Soft.
The company was founded by Dr. Wei Liu, a distinguished figure in the AI landscape, known for his previous role as a Tencent Distinguished Scientist and his recognition as an IEEE/AAAS Fellow. Dr. Liu also notably spearheaded Tencent’s Hunyuan Large Model, bringing a wealth of expertise to Video Rebirth’s innovative endeavors.
Dr. Liu articulated the company’s core mission, stating, “Video generation is the most direct pathway to building interactive world models, but the industry’s scaling laws are not yet solved.” He further elaborated on their strategic direction: “Our mission is to move beyond simple ‘text-to-video’ tools and build a true generative platform. We are engineering for the creators who need cinematic quality and physical consistency, not just novelty. This funding allows us to accelerate our research and build an ecosystem for the next AI Generated Entertainment (AIGE) trend.”
Video Rebirth is poised to disrupt the current generative video market by shifting the industry’s focus from consumer-grade tools, which often fall short in meeting professional standards, to a platform offering high-fidelity and fine-grained control. The company’s core technological advancement lies in its proprietary “Physics Native Attention” architecture. This novel approach, coupled with a meticulously curated data strategy, enables its “Bach” series of models to produce content with unparalleled realism in aspects such as light, shadow, and intricate object interactions.
Also Read:
- mimic Robotics Secures $16 Million to Advance AI-Powered Dexterous Automation
- Resolve AI Secures $35M Seed Funding to Revolutionize Software Operations with AI Production Engineers
The startup plans a significant milestone with the launch of its 1.0 product in December 2025. This platform is expected to directly compete with and challenge existing AI video generation models, including prominent offerings like OpenAI’s Sora, by delivering superior cinematic quality and physical consistency essential for professional-grade productions.
