Abstract: With the rapid growth of the software industry, reliability research on software has the biggest impact on system reliability and software quality improvement. In this study, we compared and analyzed the attributes of the software reliability model dependent on the life distribution of the Type-2 Gumble model and the Erlang Model with Goel-Okumoto Basic Model. Software failure time data were used to identify software failure phenomena and the maximum likelihood estimation method was used for parameter estimation. As a result in terms of the intensity function, the Type-2 Gumble Model showed a tendency to decrease as the failure time progressed compared to the Goel-Okumoto Basic Model indicating that it was an efficient model but the Erlang Model increased inversely. In the pattern of the mean value function, the Type-2 Gumble Model showed some overestimation pattern compared to the true value but the error width was smaller than the Erlang Model. In addition as a result of comparing reliability by applying future mission time, the Type-2 Gumble Model was stable with the Goel-Okumoto Basic Model but the Erlang Model showed a decreasing tendency. Through this study, we have newly analyzed the attributes of software reliability on Type-2 Gumble Model without existing research examples and Erlang Model widely used in reliability field and it was able to present new research information that software developers could use as basic guidelines.