Treating the bulk phase with high-level ab initio methods, such as coupled cluster, is a nontrivial task because of the computational costs of these electronic structure methods. In this part of our hydrogen fluoride study we make use of the quantum cluster equilibrium method, which employs electronic structure input of small clusters and combines it with simple statistical mechanics in order to describe condensed phase phenomena. If no parameter adjustment is applied, then the lower quantum chemical methods, such as density functional theory in conjunction with the generalized gradient approximation, provide wrong results in accordance with the description of the strength of the interaction in the clusters. While density functional theory describes the liquid phase too dense due to overbinding of the clusters, the coupled cluster method and the perturbation theory at the complete basis set limit agree well with experimental observations. If we allow the two parameters in the quantum cluster equilibrium method to vary, then these are able to compensate the overbinding, thereby leading to very good agreement with experiment. Correlated methods in combination with small basis sets giving rise to too weakly bound clusters cannot reach this accuracy even if the parameters are flexible. Only at the complete basis set limit, the performance of the correlated methods is again excellent.