With the increasing demand for water conservancy informatization, to address the problems of poor generality, low reuse rate, and update maintenance difficulty in the development of cloud model service platform for water resources planning. Based on cloud computing technology and combined with business-oriented workflows, construct a cloud model service platform for watershed water resource planning. On the basis of constructing the watershed digital water grid, this study tightly couples the water cycle model with the multi-objective optimization configuration model of water resources, and relies on the digital topological water grid to achieve dynamic configuration of regional water resources. The cloud model service platform adopts a business-oriented modeling approach, with the model library as the core, based on the B/S development architecture. Taking the Shaanxi section of the Weihe River basin as an example, the study simulates and analyzes the evolution of water cycle in the research area, and conducts multi-objective optimization configuration research on water resources at the annual planning level. The results show that: 1) the relative errors of monthly average runoff during the calibration and validation periods are − 1.67% and − 4.63%. The Nash-Sutcliffe efficiency coefficients of monthly average runoff are 0.815 and 0.795, indicating that the cloud model service platform can well depict the variation process of runoff during the validation period, respectively; 2) in the water resources allocation model of the cloud model service platform, using the NSGA-III algorithm and the multi-attribute decision-making model selects reasonable water resources allocation schemes for the Weihe River basin. The water demand in 2025 is 6.34×109 m³, the water supply is 5.84×109 m³, and the water shortage rate is 7.95%. The research results can provide technical reference and inspiration for the construction of smart water conservancy and refined allocation of water resources in the Weihe River basin.