To establish the growth prediction model for Salmonella in retail chicken，Salmonella enteritis in fresh chicken sampled from supermarket at different temperature were studied. The growth of Salmonella enteritidis in retail chicken was studied at 4，10，16，25，30 and 37 ℃，and the growth curve was drawn. The first grade model was fitted with 3 models，including Gompertz，Richards and Logistic in CurveExpert software and Baranyi model in DMFit software，and the most suitable model was determined. After fitted by four models，Baranyi model showed the best fitting with all the correlation coefficients at different temperature above 0.98，which was the most suitable for predicting the growth dynamics of Salmonella in chicken. The second level model，which described the effect of temperature on the maximum growth rate，was predicted by putting the fitting data obtained from the first level into Ratkowsky equation. The accuracy of the model was tested by the accuracy factor，the bias factor and the root mean square error. Results showed that the accuracy factor of this second growth model was 1.129 698，the bias factor was 0.984 85. the root mean square error was 0.091 5，and the coefficient of determination was 0.982 5. The results showed that the reliability of the established model was high.