<p>Balancing exploration and exploitation remains a fundamental challenge in the design of metaheuristic algorithms, as agents typically cannot perform both tasks simultaneously. To address this issue, this article introduces the filling space swarm optimization (FSSO) algorithm that explicitly segregates the population into two distinct groups—explorers and exploiters—each governed by specialized, purpose-driven mechanisms. The exploration process is guided by a virtual particle that traces a trajectory defined by a space-filling curve, ensuring systematic coverage of the entire search space. Acting as a dynamic, centralized reference, this virtual particle coordinates the movement of explorer agents, enhancing global search efficiency and avoiding redundant evaluations. In parallel, the exploitation process is driven by a new diamond crossover operator, which intensifies the search around elite solutions by generating candidate solutions within a diamond-shaped neighborhood. This operator promotes fine-grained local optimization, boosting the algorithm’s ability to refine high-quality solutions. The adaptive balance between exploration and exploitation is maintained through the adjustable ratio of explorers to exploiters, enabling the algorithm to respond dynamically to different search phases. Extensive validation using a diverse set of 30 benchmark functions, including shifted functions designed to challenge center-biased strategies, confirms the superiority of the proposed approach, by comparing it with eight well-known metaheuristics schemes. The results reveal that the algorithm achieves faster convergence and higher solution quality compared to state-of-the-art metaheuristics outperforming them in 80% of the benchmark functions, while successfully avoiding the limitations associated with center-biased search behaviors.</p>