To address the high torque issue of gas control valves during opening and closing, a multi-factor analysis and structural optimization design study is conducted, and a low-torque optimization approach integrating topology optimization, response surface methodology, and non-dominated sorting genetic algorithm II (NSGA-II) is proposed. By establishing a theoretical opening/closing torque model of the control valve, it was clarified that mechanical friction torque was the dominant influencing factor, and the coupling effect of medium-induced force, spring preload, and Glyd ring compression ratio on torque and sealing performance was analyzed in detail. In the structural optimization process, the valve seat shape was reconstructed through topology optimization to reduce the effective medium-acting area and frictional resistance. Subsequently, a multi-objective optimization model with mechanical friction torque and leakage rate as objectives was constructed based on the response surface regression model, and the torque and sealing performance were simultaneously optimized by combining the NSGA-II. The experimental results showed that under a medium pressure of 5.2 MPa, the mechanical friction torque of the optimized control valve was reduced by 71.8%, validating the accuracy and feasibility of the proposed optimization approach. The research results provide a theoretical basis for high-performance design and localization of gas control valves.