Physicl Secures $10M Seed Round to Build Realistic Simulated Worlds for Robot Training

Physicl, a startup founded by Alex de Vigan, has raised $10 million in seed funding led by Inovia Capital to help robot builders and world model designers train their systems using more realistic simulations.

De Vigan spent the past decade building nfinite, a platform that let brands and retailers create 3D models of products for online display. When transformer-based AI models pushed that company toward generative imagery, he saw a new use for the rendering technology his team had developed. "What we were super unique about doing in 3D, AI was going to replace, but something was missing in AI: its ability to understand, interact, and manipulate space and objects," de Vigan said.

Researchers working on physical AI have struggled with the limits of training robots on video and the cost of gathering data through remote operations. Simulated digital worlds offer a promising alternative, but making them faithful to real-world physics has proven difficult. De Vigan believes the pipeline nfinite built to turn messy real-world information into high-fidelity 3D representations with correct geometry, materials, and physics can now make physical AI models more effective.

Physicl is entering a crowded field of startups supplying data and tooling to foundation model builders and robot operators. "The space is very young and very blurry sometimes," de Vigan admitted. The company plans to focus on post-training models for difficult edge cases, and is releasing a product called Studio with 10,000 3D environments and 100,000 assets that lets users generate synthetic training data for specific problems, such as helping a robot distinguish a cup on a cluttered counter or in shadow.

SpAItial, a world model builder, worked with Physicl as an early customer to improve the fidelity of its products, including its Echo-3 tool released last week, which generates 3D simulations from 2D imagery. CEO and founder Matthias Niessner said training on Physicl's high-fidelity synthetic data was necessary. Physicl aims to be more than a data provider, positioning itself as tooling for ongoing synthetic data generation. "What we created 10 years ago, without knowing it, was basically an agentic workflow to transform unstructured information into high fidelity 3D," de Vigan said.

Source: TechCrunch