This study aimed to investigate the association between the plant-based diet index (PDI) score and T2D risk among residents of Shanxi Province, China, and explore whether the association was influenced by different levels of cardiovascular risk. A total of 50,694 participants aged 35–75 years were recruited between 2017 and 2019, and they were further divided into the high cardiovascular risk population (HCRP; n = 17,255) and the non-high cardiovascular risk population (non-HCRP; n = 33,439). The PDI was calculated based on food frequency from a food frequency questionnaire (FFQ). Incident T2D was defined based on elevated plasma glucose (≥7 mmol/L) or hypoglycemic medicine use. We investigated the association of the PDI andT2D risk using a two-level generalized estimating equation and restricted cubic splines model. The results showed that quartile 4 of the PDI indicated significantly reduced T2D risk in the total population (OR: 0.83; 95% CI: 0.75–0.92), HCRP (OR: 0.80; 95% CI: 0.71–0.91), and non-HCRP (OR: 0.80; 95% CI: 0.74–0.87) compared with corresponding quartile 1 (OR = 1). In stratified analysis, the negative associations between PDI and T2D risk were stronger in the total population with the elderly (age > 60 years), BMI < 24, and men, and in the non-HCRP with men and BMI 24–28, and in the HCRP with the elderly and BMI < 24 than those with corresponding subgroups (pinteraction < 0.05). Linear curves were observed for the total population and non-HCRP, but an L-shaped association was observed for the HCRP. Therefore, our results suggest that higher PDI scores may effectively attenuate the T2D risk in the Chinese population and non-HCRP, and a beneficial association of PDI with T2D risk was observed in the HCRP at a certain threshold level. Longitudinal studies and intervention trials are required to validate our study findings.