To establish a population pharmacokinetics (PopPK) model of nirmatrelvir in Chinese COVID-19 patients and provide reference for refining the dosing strategy of nirmatrelvir in patients confirmed to be infected with SARS-CoV-2. A total of 80 blood samples were obtained from 35 mild moderate COVID-19 patients who were orally administered nirmatrelvir/ritonavir tablets. The PopPK model of nirmatrelvir was developed using a nonlinear mixed effects modeling approach. The stability and prediction of the final model were assessed through a combination of goodness-of-fit and bootstrap method. The exposure of nirmatrelvir across various clinical scenarios was simulated using Monte Carlo simulations. The pharmacokinetics of nirmatrelvir were well characterized by a one-compartment model with first-order absorption, and with creatinine clearance (Ccr) as the significant covariate. Typical population parameter estimates of apparent clearance and distribution volume for a patient with a Ccr of 95.5 mL·min-1were 3.45 L·h-1 and 48.71 L, respectively. The bootstrap and visual predictive check procedures demonstrated satisfactory predictive performance and robustness of the final model. The final model was capable of offering an early prediction of drug concentration ranges for different nirmatrelvir dosing regimens and optimize the dose regimen of nirmatrelvir in individuals with confirmed SARS-CoV-2 infection.