Algorithms for optimization are crucial for solving complex numerical problems in many scientific and technical domains. In this paper local and global search, respectively, metaheuristic techniques Particle Swarm Optimization (PSO) and the Grasshopper Optimization Algorithm (GOA) have demonstrated significant potential. Every technique has drawbacks, though; PSO might not be able to explore high-dimensional search spaces, whereas GOA frequently suffers from premature convergence. In order to improve optimization performance, this study suggests a hybridized strategy that combines GOA's exploratory nature with PSO's exploitative capabilities. The two techniques are dynamically balanced by the hybrid model, which improves convergence and resilience when dealing with high-dimensional and multimodal optimization issues. Benchmark functions are used for performance evaluation, and the results show notable gains in convergence time, solution accuracy, and stability when compared to solo GOA, PSO, and other traditional optimization methods. For numerical optimization problems in the real world, such as financial modelling, machine learning parameter tweaking, and engineering design, the suggested hybrid technique presents a viable answer