- SpAItial, led by AI 3D pioneer Matthias Niessner, raised $13M to build text-to-interactive-3D foundation models.
- The team includes top talent from Google and Meta, aiming to enable fully interactive, photorealistic 3D worlds.
- SpAItial will license its models to developers, targeting gaming, AR, robotics, and digital twin applications.
Matthias Niessner, renowned for his work in AI-driven 3D modelling, has launched SpAItial, a startup with the ambitious goal of creating foundation models capable of generating interactive, photorealistic 3D environments from simple text prompts.
SpAItial’s $13 million seed round—led by Earlybird Venture Capital with Speedinvest and notable angels—marks one of Europe’s largest early-stage raises in the space.
Niessner, a former Synthesia cofounder and professor at TU Munich, has assembled a powerhouse team including ex-Google 3D teleconferencing expert Ricardo Martin-Brualla and Meta’s former text-to-3D lead David Novotny.
The startup’s teaser video hints at the potential: a text prompt instantly creates a detailed 3D room.
But Niessner’s vision goes further—he wants these worlds to be fully interactive, behaving like real environments where users can manipulate objects, opening up applications from gaming to digital twins and robotics.
While competitors like Odyssey and World Labs are also pursuing photorealistic 3D generation, Niessner believes the market is still wide open, with the true “holy grail” being seamless, user-driven creation of entire worlds.
SpAItial plans to license its foundation model to developers, enabling a range of downstream applications, and is prioritising strategic partnerships and early revenue.
The startup will use its funding for hiring top talent and scaling compute resources, focusing on quality over headcount.
If successful, SpAItial could transform how games, AR/VR, and even real-world simulations are built, potentially letting anyone create their own interactive 3D universe with a few words.
Edited by Annette George
