The isolated gas-phase actinide dioxide dications, AnO 2 2+ , were evaluated by DFT and coupled cluster CCSD(T) calculations for 12 actinides, An = U−Lr. CASSCF calculations were used to define the orbitals for the CCSD(T) calculations. The characteristic linear [OAnO] 2+ hexavalent actinyl(VI) was found to be the lowest energy structure for An = U, Np, and Pu, which also form stable actinyl(VI) species in solution and possibly for Am when spin−orbit effects are included. For Am, there is a divalent [An II (O 2 )] 2+ structure where the dioxygen is an end-on physisorbed η 1 -3 O 2 2 kcal/mol above the actinyl when spin−orbit effects are included which lower the energy of the actinyl structure. For An = Cm, Bk, and Lr, the lowest energy structure is trivalentwhere the dioxygen is a side-on superoxide, η 2 -O 2 − . For Cm, the actinyl is close in energy to the ground state when spin−orbit effects are included. For An = Cf, Es, Fm, Md, and No, the lowest energy structure is divalent [An II (O 2 )] 2+ where the dioxygen is an end-on physisorbed η 1 -3 O 2 . The relative energies suggest that curyl(VI) and berkelyl(VI), like well-known americyl(VI), might be stabilized by coordinating ligands in condensed phases. The results further indicate that for californyl and beyond, the actinyl(VI) moieties will probably be elusive even using strong donor ligands. The prevalence of low oxidation states (OSs) An(II) and An(III) for transplutonium actinides reflects stabilization of the 5f orbitals and validates established trends, including the remarkably high stability of divalent No. Bond distances and other parameters suggest maximum bond covalency around plutonyl(VI), with a particularly substantial decrease in bond strength between americyl(VI) and curyl(VI).