A tumor is a complex cluster with many types of cells in the microenvironment that help it grow. Macrophages, immune cells whose main role is to maintain body homeostasis, represent in the majority of cancers the most important cell population. In this context, they are called tumor-associated macrophages (TAMs) because of their phenotype, which contributes to tumor growth. In order to limit the use of animals, there is a real demand for the creation of in vitro models able to represent more specifically the complexity of the tumor microenvironment (TME) in order to characterize it and evaluate new treatments. However, the two-dimensional (2D) culture, which has been used for a long time, has shown many limitations, especially in terms of tumor representation. The three-dimensional (3D) models, developed over the last 20 years, have made it possible to get closer to what happens in vivo in terms of phenotypic and functional characteristics. Due to their architectural similarity to in vivo tissues, they provide a more physiologically relevant in vitro system. Most recently, it is the development of 3D coculture models in which two or three cell lines are cultured together that has allowed a better representation of TME with cell–cell interactions. Unfortunately, there is no clear direction for the design of these models at this time. In this Review on the coculture between cancer cells and TAMs, an in-depth analysis is performed to answer multiple questions on the conception of these models: Which models to use, and with which material and cancer lineage? Which monocyte or macrophage lines should be added to the coculture? And how can these models be exploited?