Detecting and quantifying quantum entanglement remain significant challenges in the noisy intermediate-scale quantum (NISQ) era. This study presents the implementation of quantum support vector machines (QSVMs) on IBM quantum devices to identify and classify entangled states. By employing quantum variational circuits, the proposed framework achieves a runtime complexity of \(O(\frac{N t}{\epsilon ^2})\) , where N is the number of qubits, t is the number of iterations, and \(\epsilon\) is the acceptable error margin. We investigate various quantum circuits with multiple blocks and obtain the accuracy of QSVM as measures of expressibility and entangling capability. Our results demonstrate that the QSVM framework achieves over 90% accuracy in distinguishing entangled states, despite hardware noise such as decoherence and gate errors. Benchmarks across superconducting qubit platforms (e.g., IBM Perth, Lagos, and Nairobi) highlight the robustness of the model. Furthermore, the QSVM framework effectively classifies two-qubit states and extends its predictive capabilities to three-qubit entangled states. This work marks a significant advancement in quantum machine learning for entanglement detection.
