The present study investigates the influence of initial geometric imperfections on large amplitude vibration of shallow sandwich shells with functionally graded graphene nanoplatelets reinforced porous (FG-GNPRP) core embedded between two aluminium facets. The aluminium core is reinforced with graphene nano-platelets (GNP) and the effective properties such as elastic modulus, and mass density, etc. are estimated using Halpin-Tsai and Voigt models, respectively. The continuous functions are adopted to accomplish micro-structural gradation while creating pores which imparts the variation in material properties along thickness direction. The nonlinear governing equations based on higher order shear deformation theory and Sander’s approximation are derived using variational approach. The nonlinearity is introduced in von-Kármán sense and various imperfection modes such as sine, global and local types are considered in transverse direction. The nonlinear results are accessed using finite element method in conjunction with direct iterative technique. The influence of porosity, GNP weight fraction, imperfection amplitude, thickness, side, and radius ratios on the vibration response of the sandwich shells is explored in detail.