Ising machines have recently been attracting attention due to their apparent ability to solve difficult combinatorial problems using analog operational principles. Oscillator Ising Machines (OIM) are especially attractive because they can be implemented easily as integrated circuits (ICs) in standard CMOS electronics. We explore the performance of OIM for decoding noisy Multi-User MIMO signals, a problem of considerable interest in modern telecommunications. Our results indicate that OIM-based decoding achieves error rates almost as good as the optimal Maximum Likelihood method, over a wide range of practical signal-to-noise (SNR) values. At high SNR values, OIM achieves ~20x fewer errors than LMMSE, a decoding method used widely in industry today. We also investigate the influence of parameter precision on decoding performance, finding that using 6 or more bits of precision largely retains OIM’s advantages across all SNR values. We estimate that straightforward CMOS OIM implementations can easily solve MU-MIMO decoding problems in under 10ns, more than 100x faster than current industrial requirements. We conclude that oscillator Ising machines can be effective for real-world applications, possibly serving as an important enabler for future telecommunication standards. Our results and data provide guidance for designing hardware OIM prototypes specialized for MU-MIMO decoding.