Dynamic load balance is the critical pivotal role of network parallel and distributed computing. In order to solve the drawbacks of BP neural network, RBF neural network(RBFNN) is applied to dynamic load balance of network. And genetic algorithm is introduced and tried in optimizing the parameters of RBF neural network, the method is well suited for searching global optimal values. In the paper, genetic algorithm and RBF neural network (GA-RBFNN) is adopted to dynamic load balance of network. The cases are applied to study the ability of dynamic load balance. The experimental results indicate that GA- RBF neural network is better dynamic load balance method than BP neural network.