Cloud load balancing is crucial for ensuring efficient use of resources and quality of service in modern cloud computing. However, achieving an optimal balance of workloads across servers is challenging due to the dynamic, large-scale, and NP-hard nature of the task scheduling problem. In this paper, a novel task scheduling optimization algorithm based on hunting behavior of the Aquila (eagle). The algorithm is modeled on four strategic phases: global exploration; vertical stoop to identify a broad search space; divergent exploration to discover diverse scheduling solutions; convergent exploitation to refine promising task resource mappings; and final exploitation to capture the optimal scheduling configuration. The Aquila-inspired approach balances exploration and exploitation to avoid local optima and improve scheduling efficiency. Experimental results show efficient load distribution, shortening runtime, and better energy consumption compared to existing algorithms. The performance of the Aquila Optimization (AO) algorithm improves its efficiency to 96.01%.