MQC is a classical postprocessing method for improving the quality of drawn samples by quantum annealers (like the D-Wave quantum processors), and any Ising processing unit (IPU). MQC receives a sample-set and coefficients of an Ising model as input, and (in most cases) result in a new/synthetic sample that is notably better than all input samples. MQC has outperformne recent software/hardware advancements such as applying spin-reversal transforms, increased inter-sample delay between successive reads/samples, applying steepest descent method (as a postprocessing approach) to drawn samples, reverse annealing, etc. It is worth noting that MQC requires significantly fewer samples.

Python implementation of MQC