High-density polyurethane (PU) foam-filled aluminum tubes represent a significant advancement in composite systems, boasting an exceptional strength-to-weight ratio and sustainability. These materials offer an economical solution for lightweight structural applications, including seismic braces, due to their effective energy absorption capacities. Despite the demonstrated performance of high-density PU foam-filled aluminum tubes in compression and their adequate bonding capacities, there is a notable absence of analytical and empirical models specifically for PU foam as an infill material. This study aims to address this gap by comparing existing concrete-filled tube (CFT) analytical models with experimental results and developing an empirical model using machine learning algorithms for aluminum tubes filled with high-density PU foam. The findings reveal that the diameter-to-thickness (D/t) ratio and the length-to-diameter (L/D) ratio significantly influence compressive resistance, affecting local and global buckling, respectively. Additionally, the material strength and geometric area of both the infill material and the metal envelope are critical factors. The study offers valuable insights into the parameters that govern compressive strength, providing recommendations for futuristic optimal design approaches.