CP2K

CP2K is a world-class, open-source molecular dynamics and electronic structure software package designed for high-performance computing (HPC) environments. At its core, CP2K utilizes the Gaussian and Plane Waves (GPW) method, which provides a highly efficient and scalable approach to Density Functional Theory (DFT) calculations. As of 2026, it remains a dominant force in the materials science and chemical research sectors due to its ability to handle large systems with linear scaling (LS-DFT) and its robust support for heterogeneous computing architectures, including GPU acceleration via CUDA and ROCm. The software architecture is primarily written in Fortran 2008, designed for massive parallelism using MPI, OpenMP, and specialized libraries like ScaLAPACK and LIBXSMM. CP2K excels in simulating complex systems, from solid-state physics and liquid-phase chemistry to biological macromolecules. Its market position is solidified by its integration into automated AI discovery workflows, where it serves as the ground-truth engine for training graph neural networks (GNNs) and other machine-learned interatomic potentials (MLIPs).