Accumulating evidence underscores the importance of cerebral fluid dynamics in maintaining brain homeostasis. The coordinated transport and exchange of substances though cerebral vasculature, mediated by the glymphatic system and the blood-brain barrier, contributes to the clearance of metabolic waste such as amyloid-beta. Dysregulation of these transport processes may underlie the development of disorders such as Alzheimer's disease and hydrocephalus. To reveal the underlying mechanisms, we developed a multiscale modeling framework for flow and mass transport simulation in cerebral vasculature. The novel vascular model consists of hierarchical networks with specially designed geometries. Vessel geometry and resulting flow characteristics can significantly influence endothelium-related biological functions. However, a major challenge for existing perfusable models lies in replicating the sophisticated hierarchical interconnections and bifurcations characteristic of native cerebrovascular systems. Therefore, we propose a growth-based method grounded in physiological principles to generate complete vascular networks encompassing the entire hierarchy—from large arteries and veins to microcirculation. Structures designed by this method simultaneously match the metabolic demands of the embedding tissue and fulfill prescribed morphological properties, providing a more biomimetic platform for studying physiological vascular dynamics. We further explored model extension with consideration of microscale fluid and mass exchange with interstitial fluid, contributing to a deeper understanding of cerebral fluid dynamics. The model was calibrated using existing multi-scale data, demonstrating consistency with both trans-scale transport measurements from rodent and human. The flow patterns revealed by this model contribute to elucidating the underlying mechanisms of cerebral fluid dynamics.