Network Bio launched with $50 million in financing and a collaboration valued at $30 million with Nvidia to develop a cell-free RNA (cfRNA) foundation model. The startup said it will train models using large-scale tissue, blood, molecular, and longitudinal clinical outcome datasets from academic medical centers including Mass General Brigham, the University of Pennsylvania, and the University of Colorado Anschutz. The company plans to use Nvidia’s Parabricks software and BioNeMo Recipes tools, and it expects to generate population-scale cfRNA representations through a model it described as Nexus, a self-supervised transformer trained on cfRNA expression profiles. Network Bio said the foundation model is intended to streamline data conversion from biobanks and support disease detection, biomarker discovery, and drug development. The announcement places additional emphasis on cfRNA as a scalable signal source for medical intelligence systems.