<p>Arid Central Asia (ACA) represents the largest non-zonal arid region in the Northern Hemisphere, and its climate evolution has profound implications for regional water resources and ecological environments. Investigating the climate evolution of the western and eastern parts of ACA is essential for improving our understanding of the spatially heterogeneous climate responses across this region. In this study, a 150-kyr transient simulation performed with an isotope-enabled model from the CRESTS project was employed to explore the orbital-scale spatial differences in climate evolution between the eastern and western parts of ACA, with a focus on variations in precipitation oxygen isotopes (δ¹⁸O). The results show that the variations of annual precipitation δ¹⁸O in both CA in the western part of ACA and the Xinjiang region of China (Xinjiang) in eastern ACA exhibit significant precession cycles. Notably, the annual precipitation δ¹⁸O in each region is dominated by its rainy-season precipitation δ¹⁸O. Comparisons among different external forcing experiments indicate that orbital forcing is the primary driver of rainy-season precipitation δ¹⁸O variations in both regions, and the rainy-season precipitation δ¹⁸O is mainly controlled by rainy-season surface temperature. Further analyses reveal marked differences in the dominant cycle of orbital-scale rainy-season precipitation between these two regions. The dominant cycle of rainy-season precipitation in CA is ∼23 kyr, while that in the Xinjiang region exhibits significant ∼41 and ∼100-kyr cycles. This discrepancy is closely associated with the differences in the cycles of 500 hPa westerly wind intensity between these two regions. Overall, this study reveals the climatic significance of orbital-scale precipitation δ¹⁸O in the western and eastern parts of ACA, and also contributes to improving the understanding of spatial differences in orbital-scale climate evolution between these two regions.</p>