Chemical reactor networks (CRN) models were developed for lean premixed gas turbine combustor to predict the NOx emissions. In this study, CRN models are constructed based on the computational fluid dynamics (CFD) for both non-pilot and pilot flame cases. Predictions of NOx emissions in combustor with the developed models were made by using CHEMKIN code and full GRI 3.0 chemical kinetic mechanism in the CRN. The predicted results agree reasonably well with the experimental data obtained from a simplified test combustor for the GE7FA gas turbine. The effects of overall equivalence ratio, swirl angle and pilot fuel ratio on the NOx emissions were investigated.