Scaling of variational quantum eigensolver performance in a superconducting quantum processor

The calculation of molecular energies and dynamics from first-principles is a classically intractable problem due to the exponential scaling in computational cost with both the number of atoms and the basis set size. Variational quantum-classical hybrid algorithms offer a promising route towards solving currently intractable molecular structure problems in near-term quantum processors. We experimentally quantify the scaling of convergence, speed, and accuracy of a hybrid quantum-classical variational simulation of chemical energies using a gate-based superconducting quantum processor with up to 8 qubits.