<p>Cold spray coatings are used in the aerospace, mechanical, and automotive industries to enhance the wear resistance, corrosion resistance, and oxidation resistance of components. These properties are strongly influenced by the adhesion and pores of the coating. Therefore, it is crucial to investigate the mechanisms of interfacial adhesion and pore formation and to quantitatively evaluate adhesion strength and porosity for improving coating quality. To this end, this paper proposes dual-particle and multi-particle cold spray models using updated Lagrangian Smoothed Particle Hydrodynamics. These models incorporate a physically based interaction model to characterize metallurgical bonding and utilize multi-GPU parallel computing to enhance computational efficiency. The proposed model is first validated using two benchmarks and experimental results from single-particle deposition. Next, the dual-particle model is employed to analyze the mechanisms by which three typical interparticle interactions affect interfacial adhesion and pore formation. Finally, the multi-particle model is used to quantitatively evaluate the adhesion strength and porosity of the deposit, and its predictive capability is validated against existing experimental data. Using this evaluation method, the effects of particle diameter distribution, spatial density, and impact velocity on these properties are investigated. This computational framework provides a reliable tool for optimizing cold spray parameters, thereby facilitating the expansion of cold spray applications in industrial fields.</p>