In order to establish the growth model of Staphylococcus aureus，the frozen surimi product was inoculated with S. aureus and Gompertz model, Logistic model, Richards model, etc, were selected to fitting the growth data of S. aureus at 10, 15, 20 and 25 ℃, respectively. Residual distribution and degree of fitting (RSS, RSE, AIC) analyses were used to evaluate model fitting ability and choose the optimal model for the growth of S. aureus in frozen surimi. The results of the comparison showed that the smallest RSS, RSE and AIC values fitted by Logistic model at 10 ℃ were 0.742, 0.352 and -4.458, respectively. The smallest RSS, RSE and AIC values achieved by Gompertz model were 0.403，0.259，-9.961；0.285，0.239，-13.089 and 0.273，0.234，-13.473, respectively. Compared comprehensively, the optimal growth model of S. aureus in the frozen Surimi was Logistic and Gompertz model was appropriate for 10 ℃ and 15 ℃, 20 ℃, 25 ℃, respectively.