Mainstream applications of density functional theory or wavefunction theory rely on benchmark studies to establish the accuracy of (approximate) methods. Primarily, such benchmarks rank accuracy by comparing total/relative energies. Energy is a convenient and obvious metric to establish method performance.
By contrast the electron density and related quantities (e.g. dipole moments, static polarisabilities) are far less prevalent in such studies. With good reason; ‘good’ electron densities are both more computationally expensive to obtain and harder to perform analyses on than traditional energy-based metrics. Resultantly, there is a disparity in literature between the density vs energy benchmark realms.
Herein we aim to close the gap by presenting a modern dataset that can also serve as template for future electron density benchmarking.