The inefficiency of the replication process in HIV, like in any retrovirus, gives rise to many variants. The observed variability reflects both viability of the mutant and selection pressures from the immune system. I will review some recently developed methodology to study various anspects of the molecular evolution of HIV: to quantify heterogeneity, to compare subgroups, to detect correlated mutations and incorporate them into phylogenetic reconstructions, and to link the sequence information to specific biological characteristics. Finally, I will relay the results of an on-going project where we have gathered a large number of sets of sequences from the HIV literature to select the the best-fitting evolutionary model for different regions of the HIV genome.