SpAItial is developing foundation models that can generate full, coherent and interactive 3D environments from text prompts, aiming for photorealism and realistic physics (e.g., breakable glass). The founding team is led by Matthias Niessner, who took an entrepreneurial leave from his visual computing and AI lab at the Technical University of Munich and previously co-founded Synthesia, and includes Ricardo Martin-Brualla (formerly on Google’s Beam project), David Novotny (led text-to-3D asset work at Meta) and Luke Rogers on the business side. The company has released a teaser demonstrating text-to-3D room generation but has limited public product beyond that. SpAItial plans to license its foundation model to developers via APIs and to work with early partners to build downstream applications. Management says it will prioritize revenue early while also investing in compute and selective hiring focused on quality rather than rapid headcount growth. Potential use cases cited include video game creation, entertainment, construction visualizations, digital twins and robotic training, and the team sees competitors such as Odyssey and World Labs in the space.

Total raised

$13M

Funding rounds

1

Latest round

Seed

Latest activity

May 2025