The design of two-stage helical gearboxes often requires balancing compactness and performance, especially in applications demanding high efficiency and limited installation space. This paper presents a multi-objective optimization framework based on the MOEA/D-DE (Multi-Objective Evolutionary Algorithm based on Decomposition with Differential Evolution) algorithm to simultaneously minimize the gearbox’s cross-sectional area and maximize its mechanical efficiency. The gearbox model considers geometric, kinematic, and strength constraints derived from standard design equations. MOEA/D-DE is adopted due to its capability to handle complex trade-offs and maintain solution diversity along the Pareto front. Numerical experiments demonstrate that the proposed method effectively explores the design space and provides a set of well-distributed Pareto-optimal solutions, offering valuable insights for gearbox designers seeking compact and high-efficiency configurations.