Purpose of reviewThe models currently proposed for the genetic architecture of obesity are critically reviewed in the light of recent developments in the search for genetic causes of this condition.Recent findingsAs for many other ‘complex’ diseases, most of the genetic variants predisposing to human obesity have not yet been identified. To date, investigation of the genetic contribution to obesity has been conducted according to two main hypotheses. The common disease/common variant hypothesis proposes that the genetic architecture of complex diseases (including obesity) is likely to consist of a limited number of alleles, each conferring a small increase in risk to the individual. Alternatively, it has also been proposed that complex diseases such as obesity may instead result from the effects of a large number of rare variants, with substantial allelic heterogeneity at disease‐causing loci. These two hypotheses have shaped strategies for the identification of disease genes, including the use of linkage analysis, association studies and the systematic sequencing of candidate genes. Linkage studies have recently been very successful in identifying new genes in which mutations cause rare monogenic syndromes of obesity. In common obesity, numerous linkage and association studies have suggested that an increasing number of genetic loci could be involved. Overall, however, these studies have failed to identify the causal genetic variants. In contrast, the direct sequencing of well‐chosen candidate genes has led to the identification of numerous rare alleles causing both syndromic and common obesity, which are less severe forms of the condition in humans.SummaryThe genetic architecture of obesity is still a matter of debate. The previously accepted hypothesis of a small number of common variants has been undermined by the low reproducibility of association studies and inconsistencies among genome scans for obesity. While high‐throughput association mapping of candidate regions holds some promise for the identification of common susceptibility alleles, it must also be considered that the genetic predisposition to obesity may instead result from multiple rare variants in a large number of genes.