Rivaroxaban is a novel oral anticoagulant widely used for thromboprophylaxis in patients with non-valvular atrial fibrillation (NVAF). The present study aimed to develop a population pharmacokinetic (PPK) model for rivaroxaban in Chinese patients with NVAF.We performed a prospective multicenter study. The plasma concentration of rivaroxaban was directly detected by high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) and indirectly by rivaroxaban-calibrated chromogenic anti-Xa assay (STAA total of 150 patients from five centers were recruited, including 263 plasma concentrations detected by HPLC-MS/MS, 2626 gene polymorphisms, and 131 plasma concentrations detected by anti-Xa assay. In our study, an oral one-compartment model was used to describe the pharmacokinetics of rivaroxaban in patients with NVAF. In the final model, the estimated apparent clearance (CL/F) and volume of distribution (V/F) were 5.79 L/h (relative standard error [RSE] 4.4%) and 51.5 L (RSE 5.0%), respectively. Covariates in the final model included creatinine clearance, total bilirubin, rs4728709, and body weight. The simulation results showed that in the 15 mg once-daily dosing regimen, in most instances the maximum plasma concentration at steady state (COur study was the first multicenter PPK model for rivaroxaban in Chinese patients with NVAF (Alfalfa-RIVAAF-PPK). The study found that 15 mg once daily may be suitable as the principal rivaroxaban dose for Chinese patients with NVAF. For patients with the rs4728709 mutation, it may be necessary to examine insufficient anticoagulation. We found that the rivaroxaban-calibrated chromogenic anti-Xa assay and HPLC-MS/MS method were highly linearly correlated. Prospective studies with larger sample sizes and real-world studies are needed for further verification.