Computational plasma physics presents unique ? ?challenges due to the extreme scales and diverse physical phenomena ? ?involved in various astrophysical and laboratory plasmas. To study ? ?plasmas in different regimes (i.e., fluid vs. kinetic) at various ? ?scales, a range of numerical models with different physics ? ?capabilities, such as magnetohydrodynamic (MHD), multi-moment, ? ?hybrid, particle-in-cell, and VlasovMaxwell codes, have been developed.

Starting with the MHD model, I will delineate the ? ?world’s largest turbulence simulation (with unprecedentedly large ? ?magnetic Reynolds number) to reveal how rapid reconnection of ? ?magnetic field lines changes the classical paradigm of the turbulent ? ?energy cascade. Additionally, I will present two massive simulations ? ?examining Rayleigh-Taylor instability (RTI) driven turbulent dynamo ? ?and laserproduced turbulence in the MHD regimes. Before ? ?transitioning to fully kinetic models, I will introduce a novel ? ?kinetic fluid model that can be used to address collisionless ? ?magnetic reconnection. Leveraging modern machine learning ? ?techniques, the accuracy of this model has been significantly ? ?improved by integrating kinetic physics from Vlasov-Maxwell ? ?simulations. Furthermore, I will briefly show a few examples using ? ?fully kinetic particle-in-cell codes. ?