NVIDIA nvmath-python is a library that bridges the Python scientific community and NVIDIA CUDA-X math libraries, providing Python users access to CUDA-X performance for common math operations. Depending on the API, operations can run on a CPU, CUDA-enabled GPU, or distributed multi-GPU, multi-node systems. With the general availability of nvmath-python v1.0, the library offers a Pythonic abstraction layer over CUDA and NVPL math libraries such as cuFFT, cuBLASLt, cuDSS, cuSPARSE, cuTENSOR, cuBLASMp, and more. A novel approach to sparsity, the universal sparse tensor (UST), enables users to create their own application-optimal sparse format through a domain-specific language without implementing it in code. Installation is fast and customizable, with options for package managers like pip, conda, uv, or pixi, and choices for CPU backend, device APIs, distributed APIs, and companion array libraries such as NumPy, CuPy, or PyTorch. nvmath-python implements core numerical operations but is not intended to replace general-purpose array libraries, focusing instead on exposing CUDA-X math library functionality in Python. It consumes and returns NumPy arrays in examples, and supports GPU libraries like cuBLAS and cuFFT, CPU libraries like NVPL for NVIDIA Grace or ARM v8 CPUs and Intel MKL for x86 hosts, and distributed libraries