A classical Chinese Natural Language Understanding (NLU) architecture usually includes several NLU components which are executed with some mechanism. A new Multilayer Search Mechanism (MSM) which integrates and quantifies these components into a uniform multilayer treelike architecture is presented in this paper. The mechanism gets the optimal result with search algorithms. The components in MSM affect each other. At last, the performance of each component is enhanced. We built a practical system–CUP (Chinese Understanding Platform) based on MSM with three layers. By the experiments on Word Segmentation, a better performance was achieved. In theory the normal cascade and feedback mechanism are just some special cases of MSM.