Diagnostic approaches based on PCR methods are increasingly used in the field of parasitology, particularly to detect Cryptosporidium. Consequently, many different PCR methods are available, both “in-house” and commercial methods. The aim of this study was to compare the performance of eight PCR methods, four “in-house” and four commercial methods, to detect Cryptosporidium species. On the same DNA extracts, performance was evaluated regarding the limit of detection for both C. parvum and C. hominis specificity and the ability to detect rare species implicated in human infection. Results showed variations in terms of performance. The best performance was observed with the FTD® Stool parasites method, which detected C. parvum and C. hominis with a limit of detection of 1 and 10 oocysts/gram of stool respectively; all rare species tested were detected (C. cuniculus, C. meleagridis, C. felis, C. chipmunk, and C. ubiquitum), and no cross-reaction was observed. In addition, no cross-reactivity was observed with other enteric pathogens. However, commercial methods were unable to differentiate Cryptosporidium species, and generally, we recommend testing each DNA extract in at least triplicate to optimize the limit of detection.