Nature Will Releases First Physical World Foundation Model: Natura-1

Nature Will has released its first foundational model for the physical world, Natura-1. The model was trained on one million hours of first-person videos showing humans handling various scenes and objects, enabling it to process unseen objects and tasks. Natura-1 supports fine-grained instructions specifying an object's final position and orientation, maintains control when grasping methods change mid-action, executes different tasks with both hands, and tracks objects that have been moved or opened throughout long-horizon tasks to determine the next steps. This announcement supplements earlier breaking news regarding Nature Will's release of the 50-billion parameter MoE model Natura-1, focusing on its physical-world capabilities.