The paper introduces a non-parametric graph clustering (NpGC) framework for multi-view data, addressing the limitations of existing methods that often require hyper-parameters and suffer from complex solving procedures. NpGC aims to eliminate hyper-parameters by using two types of anchors—view-related and view-unrelated—to concurrently mine exclusive and common characteristics among views. These anchors are gathered into a consensus bipartite graph, which extracts both complementary and consistent multi-view features, leading to superior clustering results. The framework is designed to handle datasets with over 120,000 samples efficiently due to its linear complexity. Experiments on various datasets demonstrate NpGC's strong performance compared to 17 classical approaches, showing its effectiveness and scalability. The paper also includes a detailed solution to the optimization problem and an analysis of the algorithm's convergence, space, and time complexities.The paper introduces a non-parametric graph clustering (NpGC) framework for multi-view data, addressing the limitations of existing methods that often require hyper-parameters and suffer from complex solving procedures. NpGC aims to eliminate hyper-parameters by using two types of anchors—view-related and view-unrelated—to concurrently mine exclusive and common characteristics among views. These anchors are gathered into a consensus bipartite graph, which extracts both complementary and consistent multi-view features, leading to superior clustering results. The framework is designed to handle datasets with over 120,000 samples efficiently due to its linear complexity. Experiments on various datasets demonstrate NpGC's strong performance compared to 17 classical approaches, showing its effectiveness and scalability. The paper also includes a detailed solution to the optimization problem and an analysis of the algorithm's convergence, space, and time complexities.