This ability to use quantum computing to achieve results beyond the reach of classical computing is called 'supremacy' and has been the theoretical advantage of quantum computing since research into the topic began. Google's results were reported in Nature.
The Sycamore quantum computer is fully programmable and can run general-purpose quantum algorithms. After the test and benchmarking phase of development reported here the team has been preparing applications in the areas of quantum physics simulation and quantum chemistry, as well as in machine learning.
The Nature paper relates that Sycamore, designed and fabricated by Google, that comprises a two-dimensional array of 54 transmon qubits, with each qubit tunably coupled to its four nearest neighbours. The qubits are arranged in a 9 by 6 staggered array.
One of the breakthrough developments in this design of qubit assembly is the adjustable coupler that controls connection to each nearest neighbour, the paper states.
The qubits are made of supercooled Josephson Junctions with aluminium for metallization and indium for bump-bonds between two silicon wafers. The chip is wire-bonded to a superconducting circuit board and operated at below 20mK. The processor is connected to room-temperature equipment that synthesizes control signals and reads the qubits status.
Next: What next?