An approach for Soft-output K-Best MIMO detection comprises computing an estimated symbol vector and Log-Likelihood Ratio (LLR) values for transmitted bits. The approach includes a relevant discarded paths selection process, a last-stage on-demand expansion process, and a relaxed LLR computation process. The relevant discarded paths selection process includes analyzing the K-Best paths and discarded paths at each intermediate tree level and selecting only those discarded paths for further processing that will help in LLR computation for at least one of the transmitted bits. The last-stage on-demand expansion process includes expanding K paths at the tree level 2NT−1 (NT=number of transmit antennas) on-demand to only 2K−1 lowest Partial Euclidean Distance (PED) paths at last tree level 2NT. The relaxed LLR computation scheme includes approximating LLR computations by assuming that discarded path PED is greater than or equal K-Best path PED.