Background Candida species are considered as the cause of one of the most important opportunistic fungal diseases. Accurate identification of Candida species is important because of antifungal susceptibility patterns are different among these species, so proper identification helps in the selection of antifungal drugs for the prevention and treatment. Phenotypic methods for identification of Candida species, which are widely used in clinical microbiology laboratories, have some limitations. Real‐time PCR followed by the high‐resolution melting analysis (HRMA) is a novel approach for the rapid recognition of pathogenic fungi. Molecular phylogeny is essential for obtaining a better understanding of the evolution of the genus Candida and the identification of the relative degree of the Candida species. The purpose of this study was molecular identification of Candida isolates by Real‐time PCR‐high‐resolution melting analysis and investigation of the genetic diversity of Candida species. Methods Two hundred and thirty‐two Candida isolates including 111 Candida isolates obtained from 96 HIV/AIDS patients and 121 Candida isolates obtained from 98 non‐HIV persons were identified by real‐time PCR and high‐resolution melting curve analysis. To evaluate genetic diversity and relationships among Candida species, PCR products of nine clinical Candida isolates, as a representative of each kind of species, were randomly selected for DNA sequence analysis. Results In HIV/AIDS patients, six species of Candida spp. were identified as follows: C albicans (n = 64; 57.7%), C glabrata (n = 31; 27.92%), C parapsilosis (n = 9; 8.1%), C tropicalis (n = 4; 3.6%), C krusei (n = 2; 1.8%), and C kefyr (n = 1; 0.90%). In non‐HIV persons, we identified eight species of Candida including C albicans (n = 46; 38.33%) followed by C glabrata and C krusei (each one, n = 18; 15%), C tropicalis (n = 13; 10.83%), C lusitaniae (n = 12; 5.17%), C parapsilosis (n = 10; 4.31%), and C kefyr and C guillermondii (each one, n = 2; 1.66%). Also, the phylogenetic analysis showed the presence of two main clades and six separate subclades. Accordingly, about 88.9% of the isolates were located in clade I and 11.10% of the studied isolates were in clade II. Conclusions Real‐time PCR followed by high‐resolution melting analysis (HRMA) is known as a reliable, fast, and simple approach for detection and accurate identification of Candida species, especially in clinical samples.