Abstract: A common issue in ultra-precision machine tool structures is the insufficient optimization of stiffness design and the excessive mass of structural components during the assembly phase. This issue conflicts with the increasing demands for stiffness and accuracy in ultra-precision machining equipment. To address these technical challenges, a structural design approach guided by the variable density topology optimization method was proposed. Combined with the response surface methodology and the NSGA-II genetic algorithm, multi-objective parameter optimization of key structural components was achieved. A machine tool structure design that balances lightweight construction and high stiffness was ultimately realized. Finite element simulations were conducted to compare and analyze the mass and total deformation characteristics of key components and the overall machine tool before and after optimization. The simulation results show that after topology optimization, the column mass is reduced by 6.70%, and its maximum deformation is reduced by 40.47%. The spindle box mass is reduced by 14.34%, and its maximum deformation is reduced by 22.41%. The slope of the maximum deformation curve at different positions of the overall machine tool decreases by 25.86%, and the maximum deformation at the intermediate position is reduced by 15.00%. These research findings provide significant theoretical references and engineering guidance for the structural optimization design of large-scale ultra-precision machine tools.