[Objective] Based on suction-controlled triaxial isotropic compression tests on compacted loess， this study establishes a soil-water characteristic curve （SWCC） and hydraulic conductivity prediction model that accounts for compression deformation. This work aims to provide an accurate and convenient method for predicting the SWCC and hydraulic conductivity of compacted loess under compressive deformation， facilitating seepage analysis in loess embankment engineering. [Method] This study takes the Q₂ compacted loess samples from the Phase I Project of mountain flattening and city construction in Yan’an New Area as the research object. Four sets of isotropic compression tests under different constant suctions （y=25， 37， 50， 100 kPa） were conducted by using a GDS unsaturated triaxial apparatus. Building upon the test findings， a power function related to the void ratio is introduced into the four-parameter Fredlund-Xing model to establish a SWCC model considering compressive deformation. The model has fewer parameters， requiring only 7 parameters； the tests are simple， as the model parameters can be obtained only through suction-controlled compression tests and saturated permeability tests； （3） the model also considers the influence of adsorbed water on the hydraulic conductivity. The proposed model is applied to predict the SWCC and hydraulic conductivity curve of Yan’an compacted loess within a wide suction range， and the predicted results are compared with measured test data. [Results] （1） Compressive deformation causes a significant mechanical wetting effect， specifically manifested as follows： a. during the constant suction loading process， the degree of saturation of the samples increases significantly with the development of compressive deformation； b. in the process of compressive deformation， the SWCC of the samples shifts overall to the right as the void ratio decreases， and the air entry value increases significantly. （2） The proposed model can accurately predict the SWCC and hydraulic conductivity of compacted loess under any void ratio within a wide suction range， and its prediction accuracy in the high suction range is significantly higher than that of the Gallipoli model and the CCG model. In addition， the proposed model is used to predict the influence of compressive deformation. During compressive deformation process， as void ratio decreases， the soil-water characteristic curve of compacted loess gradually shifts to the upper right， and the air entry value increases significantly. In the low suction range， the hydraulic conductivity of compacted loess decreases with the decrease of void ratio， while in the medium and high suction ranges， the hydraulic conductivity increases with the decrease of void ratio. The response of the hydraulic conductivity of compacted loess to compressive deformation is essentially caused by the change of the cross-sectional area of water flow and flow paths due to the change of micro-pores. [Conclusion] The proposed model provides a reliable tool for understanding the soil-water characteristics and permeability behavior of compacted loess under compressive deformation. It not only improves the accuracy of seepage analysis in loess filling projects but also provides a theoretical basis for evaluating the stability and safety of engineering structures built on compacted loess.