The authors of the study have developed a modification of the Goldberg model, which is able to find the only accurate solution in a specialized knapsack problem. Knapsack problems are NP-complete, which implies an imbalanced increase in computational time as the complexity of the problem increases. This study investigates a way to solve such a problem by using the PyPy interpreter and applying additional mechanisms to the modified Goldberg model. The authors performed a computational experiment to investigate the resulting genetic algorithm under different input data. The results show that the use of the modified Goldberg model with modifications allows to obtain an accurate solution for the specialized knapsack problem. The use of a third-party PyPy interpreter, developed by an open community of software engineers, reduces the computational time by multiple times.