PointMamba is a deep learning research project that adapts the Mamba State Space Model architecture for 3D point cloud analysis. It processes point cloud data using a sequential state space approach instead of traditional attention mechanisms. The method handles tasks such as point cloud classification and segmentation, achieving competitive results on standard benchmarks.