Colorectal cancer (CRC) shows high incidence and mortality, in part because of the role of the tumor microenvironment, which is often viewed as an active promoter of disease progression. Macrophages are among the most abundant cells in the tumor microenvironment. These cells are generally categorized into the classically activated (M1) phenotype with inflammatory and anti-cancer properties, or the alternatively activated (M2) subtype known as tumor-associated macrophages, which promote tumor proliferation and survival. It is thought that the M1/M2 subclassification scheme is strongly influenced by cellular metabolism; however, the metabolic divergence between the subtypes has not been fully elucidated. In particular, the metabolic states of M1 and M2 macrophages induced by cancer cells remains poorly understood. Computational modeling is needed to provide a systematic understanding of macrophage metabolism, due to the intricate and complex nature of their intracellular metabolic networks. In this work, we used patient-derived transcriptomics data to generate a suite of computational models that characterize the M1- and M2-specific metabolic states. The genome-scale metabolic models represent all the cell’s known metabolic genes, reactions, and metabolites, and are simulated to estimate flow of material (flux) through the network, and thus capture the functional state of the cell. We apply the models to assess differences in the cells’ predicted metabolic network and capability, and show key and consistent differences between the M1 and M2 macrophage metabolic signatures. Furthermore, we leverage the models to identify metabolic perturbations that cause the metabolic state of M2 macrophages to more closely resemble M1 cells, thereby identifying novel immunometabolic targets that may be of clinical relevance for cancer patients. Overall, this work increases understanding of the role of macrophages in CRC and elucidate strategies to promote the anticancer metabolic state of anti-tumor macrophages.