In this paper, the statistical inference of accelerated life tests under Type-II censoring is studied for constant stress accelerated life tests. It is assumed that the lifetime at design stress has inverse Gaussian distribution. The scale parameter of the lifetime distribution at constant stress levels is assumed to be an inverse power law function of the stress level. The model parameters and the reliability function are estimated using the maximum likelihood method. Asymptotic Fisher information matrix, the asymptotic variance-covariance matrix and the confidence intervals are founded. The predictive value of the scale parameter and the reliability function under the usual conditions are obtained under Type-II censoring. Finally, some numerical illustrations by using Monte Carlo simulations are introduced to illustrate the proposed procedures.