Appointment marks a new stage for Meshy’s research as it expands from individual 3D assets to systems that understand, create, and continuously run complete digital worlds.
SILICON VALLEY, Calif., Sept. 8, 2026 /PRNewswire/ — Meshy, the AI 3D generation company, today announced that Dr. Xin Tong, one of the most influential researchers in computer graphics and 3D vision, has joined the company as Chief Scientist.
The appointment marks a new stage in Meshy’s research. The company’s focus is expanding from individual 3D assets to systems that can understand, create, and continuously run complete digital worlds. Meshy aims to make 3D a foundational medium for human creativity — easy for individual creators to pick up, powerful enough for professional production teams, and, with richer expressiveness, capable of giving rise to entirely new interactive experiences.
As Chief Scientist, Dr. Tong will be responsible for Meshy’s long-term research strategy, leading research on next-generation 3D foundation models and driving exploration and breakthroughs in multimodal world models and real-time interactive systems. He will also help build the research team and cultivate talent, supporting the translation of research into models and products that serve creators, developers, and studios worldwide.
“Few researchers have had such a profound impact on both the foundational research and the industrial applications of modern computer graphics as Dr. Xin Tong,” said Dr. Yuanming (Ethan) Hu, Founder and CEO of Meshy. “His work has helped define how computers represent, reconstruct, generate, and render the three-dimensional world. Even with the enormous industry impact he has already achieved, Dr. Tong maintains an extraordinarily deep grasp of technical detail. Having him lead our long-term research strategy is a major step in Meshy’s path toward next-generation 3D foundation models — systems that go beyond creating individual 3D assets to generating complete interactive environments.”
Dr. Xin Tong: Academic and Industry Background
Dr. Tong received his Ph.D. from Tsinghua University in 1999 and joined Microsoft Research Asia (MSRA) the same year. He went on to spend 25 years at MSRA, where he led the Internet Graphics Group and served as Partner Research Manager. Over the course of his career, Dr. Tong has published more than 190 papers, including more than 60 in ACM SIGGRAPH and ACM Transactions on Graphics. According to Google Scholar, as of September 2026 his work has been cited more than 21,000 times, with an h-index of 80.
His research covers many of the core problems of modern 3D intelligence. O-CNN, published with collaborators at SIGGRAPH 2017, introduced an efficient octree-based architecture that enables deep learning to process high-resolution 3D shapes. In recent years, research he has co-authored includes TRELLIS, a structured 3D generative model; MoGe, a single-image geometry estimation model; and RenderFormer, which performs neural rendering of triangle meshes with global illumination. Over the past two years, he has led his team through a series of explorations in video-based character performance models.
Dr. Tong has also long championed bringing graphics research into large-scale industrial use. Research from teams under his leadership has been deployed across multiple Microsoft products and platforms, including Xbox game development technologies, the 3D printing capabilities in Windows, and Direct3D.
“Computer graphics is entering a new chapter,” said Dr. Xin Tong. “Foundation models are beginning to unify geometry, appearance, structure, and semantics, and to extend further into dynamics and interaction. The greater challenge is building unified models that scale while maintaining consistency in dynamics and interaction. What drew me to Meshy is precisely its ambition to pursue this complete technical pipeline. I look forward to laying a solid scientific foundation for that future.”
Meshy is also extending its research into real-time interactive worlds, a direction now taking shape through Mora, short for Multimodal Open-world Real-time Architecture. Mora 1 is now open to the public at mora.fun. It combines coding agents, Meshy’s 3D generation technology, and real-time video models to create worlds that people can enter, interact with, and reshape.
Mora offers an early validation of a key direction Dr. Tong will help lead: bringing spatial understanding, generative capability, and real-time interaction together in a single system. Meshy believes this exploration can make AI a force that continuously creates joy and new experiences. This is precisely Meshy’s vision: AI for Fun.
If 3D generation gave imagination a shape, generative worlds will let people truly step inside it.
About Meshy
Meshy builds foundation models for AI-powered 3D generation, turning text and images into production-ready 3D assets for games, film and 3D printing. Founded by Ethan Hu, who holds a Ph.D. from MIT, Meshy serves more than 12 million registered users worldwide. In July 2026, the company raised nearly $400 million in a Series B round at a $1.5 billion valuation, the largest round to date in AI 3D. Learn more at meshy.ai.
