Ordered and disordered A56:25B43:75 samples were computationally simulated in a BCC phase, where (A) are magnetic atoms and (B) are nonmagnetic atoms. We obtained the probability distribution of the number of first magnetic neighbors for different long-range order parameters, and simulating in this way the possible hyperfine field distribution of the system. It was found that the average number of near nearest magnetic neighbors decreases quadratically with the long-range order parameter. By using the 1=2 Ising model and the Monte Carlo method with Metropolis dynamics, the magnetic behavior of the alloy was simulated as a function of dimensionless temperature µ and of the long-range order parameter. It was obtained a para-ferro phase transition when the average number of magnetic bonds per atom exceeds the value 2:37 (bonds/atom); besides the behavior of the mean hyperfine field, for µ ! 0, as a function of the number of mean near nearest neighbors was determined in the light of 1=2 Ising model.