<p>Cloud computing has transformed the IT sector by providing on-demand, scalable computing resources. Effective task scheduling in cloud environments is vital for maximizing resource efficiency, minimizing operational expenses, and boosting overall system performance. As cloud environments grow increasingly complex, multi-objective optimization has become essential for handling diverse and dynamic workloads. To improve task execution efficiency in cloud systems, numerous metaheuristic algorithms and their variations have been developed for scheduling optimization. This study employs the most recent metaheuristics beluga whale optimization (BWO) algorithm for scheduling in conjunction with a multi-objective optimization model to improve cloud system performance. Based on that premise, this study proposed an innovative method known as improved beluga whale for cloud task scheduling (IBWC), which builds upon the latest metaheuristic techniques to optimize multiple objectives simultaneously, including execution time, load, and monetary cost. The Cauchy mutation strategy, the effective producer’s search operator of sparrow search algorithms (SSA), and quasi-opposition-based learning (QOBL) are all used by IBWC to make it better at optimizing globally, converge, and be robust. Moreover, the IBWC-Scheduler provides customizable, user-customized weights for various objectives, enabling firms to tailor optimization objectives to their specific needs. The experimental results indicate that the improved IBWC method significantly outperforms earlier algorithms, delivering an 8–11% reduction in execution time, a 20–25% improvement in load balancing, and a 20–28% reduction in pricing costs. Additionally, IBWC reduces overall cost by 14–20%, solidifying its effectiveness as a robust solution for modern cloud computing challenges.</p>