CHARMM (Chemistry at HARvard Molecular Mechanics) is a versatile and widely used molecular simulation program developed over three decades, primarily for studying biological molecules such as proteins, peptides, lipids, nucleic acids, carbohydrates, and small molecule ligands in various environments. The program offers a suite of computational tools for conformational sampling, free energy estimation, molecular minimization, dynamics, and analysis. It supports a broad range of energy functions and models, including mixed quantum mechanical-molecular mechanical force fields, all-atom classical potential energy functions with explicit and implicit solvent models, and membrane models. CHARMM has been ported to various platforms and architectures, both serial and parallel. This paper provides an overview of CHARMM, focusing on developments since its initial publication in 1983, and discusses its functional multiplicity, scripting language, atomic potential energy function, and implicit solvent methods. The program's flexibility and extensibility are highlighted, along with its applications in understanding the structure, function, and properties of complex biomolecular systems.CHARMM (Chemistry at HARvard Molecular Mechanics) is a versatile and widely used molecular simulation program developed over three decades, primarily for studying biological molecules such as proteins, peptides, lipids, nucleic acids, carbohydrates, and small molecule ligands in various environments. The program offers a suite of computational tools for conformational sampling, free energy estimation, molecular minimization, dynamics, and analysis. It supports a broad range of energy functions and models, including mixed quantum mechanical-molecular mechanical force fields, all-atom classical potential energy functions with explicit and implicit solvent models, and membrane models. CHARMM has been ported to various platforms and architectures, both serial and parallel. This paper provides an overview of CHARMM, focusing on developments since its initial publication in 1983, and discusses its functional multiplicity, scripting language, atomic potential energy function, and implicit solvent methods. The program's flexibility and extensibility are highlighted, along with its applications in understanding the structure, function, and properties of complex biomolecular systems.