Gaussian-SLAM applies Gaussian Splatting, a neural scene representation technique using 3D Gaussians, to the problem of Simultaneous Localization and Mapping (SLAM). The system performs real-time tracking and dense reconstruction from RGB-D camera input, producing photorealistic 3D maps without the rendering artifacts typical of traditional approaches. It supports standard SLAM benchmarks including Replica, TUM_RGBD, ScanNet, and ScanNet++.