Temperature is a key factor affecting bacterial growth and inactivation, which often used to control the potential risk of pathogenic microorganisms. However, due to individual differences among bacterial strains, bacteria show different inactivation trends under the same temperature, which is called bacterial inactivation variability. The inactivation variability of 19 Vibrio parahaemolyticus was compared at pasteurization temperature (65℃) and cold chain temperature (10℃), data with the Weibull model were fitted, the inactivation parameters (t_R) were received, and the inactivation variability was explored among the different strains. Under the condition of 65℃, the t_R values of 19 Vibrio parahaemolyticus between 22.62-67.23 s, VPC-1 was the most heat resistant strains, VPC-10 was the least heat resistant strains, the optimum probability distribution of heat inactivation t_R values was Normal (44.82, 12.27). Under the cold inactivation condition of 10℃, the t_R values between 113.95-371.38 h, VPC-3 was the cold resistant strains, VPC-2 was the least cold resistant strains, the optimum probability distribution of cold inactivation t_R values was Loglogistic (51.45, 148.88, 4.67). The results showed that there was no significant correlation between thermal inactivation and cold inactivation of Vibrio parahaemolyticus, the inactivation variability of the bacterial strain was widely existed in Vibrio parahaemolyticus, and the present inactivation model was only based on a single strain, so it was difficult to describe the overall inactivation trend. At the same time, a random model of bacterial inactivation variability was preliminarily constructed, and the traditional inactivation parameters were replaced by the probability distribution.