Metal−organic frameworks (MOFs) enable compositional and structural diversity and allow the incorporation of chemical information within structures to explore further levels of functionality beyond traditional solid-state materials. Potential utility of MOFs ranging from energy storage, gas storage, gas separation, water capture, and catalysis to biomedical applications justifies the continued interest in this field. The challenge of designing and predicting new crystalline MOFs is addressed with success applying a geometric approach of construction supported by reticular chemistry. This approach is based on the coordination of secondary building units (SBUs), which are molecular complexes and cluster entities, and organic carboxylate ligands. It leads to the description of preferred MOF topologies and achieves robust structures and permanent porosity in MOFs. Three new MOFs, that are being synthesized and characterized, were designed by applying this methodology. The first predicted MOF is Zn4O(BPDI)3, in which each dicarboxylic molecule coordinates to two oxocentered Zn4 tetrahedra, that are the SBUs, and each SBU is coordinated to eight organic linkers, yielding a three-dimensional open framework with cubic cages. The second predicted MOF is Zr6(-O)4(-OH)4(BPDI)6. The SBUs are Zr6-octahedron, whose faces are alternatively covered by -O and -OH groups. The coordination of each dicarboxylic ligand to two SBUs and each SBU to twelve ligands yields a three-dimensional open framework with tetrahedral and octahedral cages. Triangular windows connect each octahedral cage to eight corner tetrahedral cages. The last predicted MOF is Ce6(-O)4(-OH)4(BPDI)6, which is based on the same net and, therefore, has the same topology that Zr6(-O)4(-OH)4(BPDI)6.