Abstract: Machine learned interatomic potentials (MLIPs) are reshaping computational chemistry practices because of their ability to drastically exceed the accuracy-length/time scale tradeoff. Despite this attraction, the benefits of such efficiency are only impactful when an MLIP uniquely enables insight into a target system or is broadly transferable, where models achieving the latter are seldom reported. I will present the 2nd generation of our atoms-in-molecules neural network potential (AIMNet2), capable of accurate and efficient treatment of neutral, charged and open-shell molecules and molecular crystals. The applications include conformational energies of element-organic molecules, closed-shell and radical reactions, Pd-catalyzed reactions, crystallographic refinement of protein structures, and crystal structure prediction.