Grain refinement is a useful method to achieve material strengthening following Hall-Petch relationship, especially when the grain size drops to nano-scale. However, nano-grained metals exhibit limited tension ductility and poor work-hardening ability due to the suppressed dislocations slip in confined space of smaller grains. How to break the mismatch of strength and ductility is a perplexing issue. Recently, gradient nano-grained (GNG) materials with a gradient structure in which the grain size ranges from tens of nanometers at the surface to tens of micrometers in the core have been proposed to overcome this long-standing dilemma. Constitutive modeling and simulation is crucial to understand the physical deformation mechanism controlling the strength and ductility, and to promote the microstructure optimization for industrial application. Here, we developed a dislocation mechanism based crystal plasticity model, where multiple dislocation evolution mechanisms are considered. Furthermore, damage evolution and mechanically driven grain growth during the deformation of GNG materials are incorporated to reveal the significant role of GNG layer on the extraordinary plasticity response. The developed constitutive model was implemented in crystal plastic finite element method (CPFEM), and successfully predicted the tensile mechanical behavior of GNG copper, such as yield stress, work-hardening and ductility. Meanwhile, the modeling and simulation clearly revealed underlying deformation mechanism controlling the ductility and strengthening with the detailed spatial distribution and temporal evolution of microstructure and damage. Finally, the constitutive model was tentatively employed to optimize the balance of strength and ductility of GNG copper by manipulating the microstructure of gradient region, showing incredible consistence with the existing optimization results conducted experimentally.