Purpose: Low cardiorespiratory fitness (CRF) is a major risk factor for cardiovascular disease (CVD) and a stronger predictor of CVD morbidity and mortality than established risk factors. The genetic component of CRF, quantified as peak oxygen uptake ( VO 2peak ), is estimated to be ~60%. Unfortunately, current studies on genetic markers for CRF have been limited by small sample sizes and using estimated CRF. To overcome these limitations, we performed a large-scale systematic screening for genetic variants associated with VO 2peak . Methods: A genome-wide association study was performed with BOLT-LMM including directly measured VO 2peak from 4525 participants in the HUNT3 Fitness study and 14 million single-nucleotide polymorphisms (SNP). For validation, similar analyses were performed in the United Kingdom Biobank (UKB), where CRF was assessed through a submaximal bicycle test, including ~60,000 participants and ~60 million SNP. Functional mapping and annotation of the genome-wide association study results was conducted using FUMA. Results: In HUNT, two genome-wide significant SNP associated with VO 2peak were identified in the total population, two in males, and 35 in females. Two SNP in the female population showed nominally significant association in the UKB. One of the replicated SNP is located in PIK3R5, shown to be of importance for cardiac function and CVD. Bioinformatic analyses of the total and male population revealed candidate SNP in PPP3CA, previously associated with CRF. Conclusions: We identified 38 novel SNP associated with VO 2peak in HUNT. Two SNP were nominally replicated in UKB. Several interesting genes emerged from the functional analyses, among them one previously reported to be associated with CVD and another with CRF.