The theory of quantum coherence is an important kind of quantum resource theory, and its free operations are various kinds of incoherent operations. In the single-system coherence resource theory, the maximally coherent state is the most important quantum resource state, and it can turn into a quantum state of any other pure state. However, the situation is quite different from multipartite quantum systems: not only does no-go theorem forbidding the existence of a unique maximally coherent state exist there, but almost all pure multipartite coherent states are incomparable (i.e., some incoherent operation transformations among them are almost never possible). In order to cope with this problem, we consider general coherent resource theories in which we relax the traditional incoherent operations to operations that do not create coherence. Specifically, we consider two possible theories, depending on whether resources correspond to bipartite coherence or genuinely multipartite coherent states (each subsystem is coherent): one is the theory in which bipartite coherence states are considered as a resource and the free operations are bipartite incoherent preservation and the other is the theory that involves genuinely multipartite coherent states and fully incoherent operations. These ideas come from the research by Contreras-Tejada et al. (Phys. Rev. Lett. 122 120503), where the alternative entanglement resource theories were considered through relaxing the class of local operations and classical communication (LOCC) to operations that do not create entanglement, and they considered two possible theories depending on whether resources correspond to the multipartite entangled or genuinely multipartite entangled (GME) states. Furthermore, we show that there exists meaningful partial order (i.e. each pure state is transformable to a more weakly coherent pure state) in these two theory frames. Finally, we prove that the genuine multipartite coherent resource theory has a unique maximally coherent state (i.e. it can be transformed into any other state by the allowed free operations). Our results cover a wide class of coherent resource theories due to the free operations we introduced, and the discussion is solidified by important examples, such as entanglement, superposition, asymmetry, et al. And, how to establish the relations between these two kinds of multipartite coherent states, quantum discords and entanglements is also an interesting problem.