The a16z Show · Tuesday, July 28, 2026
World Labs has developed a base generative model named Marble, capable of creating geometrically consistent 3D worlds from image or text prompts. This technology aims to facilitate AI understanding and interaction with spatial environments, which is crucial for advancements in robotics.
“When we released the first version of our generative model called Marble last winter around November, December, Cinex just signed up.”
“Marble is the code name for the base model that World Lab is being training and iterating on. The fundamental capability right now of Marble that is publicly released is to take a prompt. It can be an image, it can be a few images or a text and turn that into a geometrically consistent world that can be represented in 3D geometry, whether it's Gaussian splat or mesh.”