In recent times, cloud computing (CC) focuses on reshaping the digital epoch as clients face significant apprehensions in privacy of their data in the cloud environment. Hence, the efficient Intrusion Detection System (IDS), capable of monitoring the packets in a network and differentiating among the behaviors of benign and malicious, is deployed. In this research, the Levy Flight—Reptile Search Algorithm (LV-RSA), and support vector machine (SVM) method are proposed for intrusion detection in the cloud environment. The LV-RSA-based feature selection is performed to choose the appropriate features for classification. Next, classification is performed by the SVM classifier to classify the type of intrusions in the cloud environment. The UNSW-NB15 and Kyoto datasets are used in the research to estimate the performance of the proposed method. The evaluation metrics of accuracy, precision, recall, f1-score, and False Alarm Rate (FAR) are taken for evaluating the proposed method. The proposed LV-RSA and SVM method obtains 99.85% accuracy on UNSW-NB15 and 98.91% accuracy on Kyoto datasets, proving more efficient than the conventional method, Hybrid Intrusion Detection Model (HIDM).