Rapid identification of
Mycoplasma bovis
infections in cattle is a key factor to guide antimicrobial therapy and biosecurity measures. Recently, nanopore sequencing became an affordable diagnostic tool for both clinically relevant viruses and bacteria, but the diagnostic accuracy for
M. bovis
identification is undocumented. Therefore, in this study nanopore sequencing was compared to rapid identification of
M. bovis
with MALDI-TOF MS (RIMM), and triplex real-time PCR in a Bayesian latent class model (BLCM) for
M. bovis
in bronchoalveolar lavage fluid (BALf) obtained from calves. In practice, pooling of samples is often used to save money, but the influence on diagnostic accuracy has not been described for
M. bovis
. Therefore, a convenience sample of 17 pooled samples containing 5 individual BALf samples per farm was analyzed as well. The results of the pooled samples were compared to the individual samples, to determine sensitivity (Se) and specificity (Sp). The BLCM showed a good Se (77.3%; 95% Credible Interval: 57.8%-92.8%) and high Sp (97.4%; 91.5%-99.7%) for nanopore sequencing compared to RIMM (Se: 93.0%; 76.8%-99.5%, Sp: 91.3; 82.5%-97.0%) and real-time PCR (Se: 94.6%; 89.7%-97.7%, Sp: 86.0%; 76.1-93.6%). Se and Sp of pooled analysis for
M. bovis
were 85.7% (95% confidence interval: 59.8-111.6%) and 90.0% (71.4-108.6%%) for nanopore sequencing and 100% (100%-100%) and 88.9% (68.4-109.4%) for RIMM, respectively. In conclusion, nanopore sequencing is a rapid, reliable tool for the identification of
M. bovis
. To reduce costs and increase the chance of
M. bovis
identification, pooling of 5 samples for nanopore sequencing and RIMM is possible.