Abstract Study question To develop an age-specific percentile distribution of anti-Müllerian hormone (AMH) in patients with polycystic ovary syndrome (PCOS) measured by three different assays. Summary answer Our age-related normograms may help to interpret AMH levels in PCOS patients and facilitate the use of AMH as a diagnostic tool across age ranges. What is known already In the last years, AMH assessment has gained widespread use in the clinical settings of several conditions. However, several challenges still exist to adopt the AMH as a marker of polycystic ovary morphology (PCOM), as included in the recently updated international guideline. Although different evaluations of age- and assay-specific reference ranges have been published in the last years, these studies have mainly been conducted in normo-ovulatory or infertile women and little is known about the higher range of AMH levels, especially in women with PCOS. Study design, size, duration Retrospective single-center cross-sectional study, including 2,725 women with PCOS who consulted the Reproductive Endocrinology and Infertility Clinic of the Erasmus University Medical Center, Rotterdam, The Netherlands. Participants/materials, setting, methods A total of 2,725 women aged 20 to 40 years with PCOS were included. PCOS was diagnosed according to the Rotterdam Criteria. Serum AMH levels were measured using the following AMH assays: (1) Gen II AMH ELISA, 2) picoAMH assay, and 3) Elecsys AMH plus assay. Age-specific centile curves were constructed using the LMS method. The correlations between AMH with clinical, endocrine, and ultrasound features were also evaluated. Main results and the role of chance To our knowledge, this is the first study with real-world data reporting AMH age-specific percentile nomograms in a large population of PCOS women using three different assays. The 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles were calculated according to a population/based approach per age (years) for the Gen II, the picoAMH, and the Elecsys assays. We also considered the age-percentile values of AMH in PCOS patients for the Gen II, the picoAMH and the Elecsys assays according to the cut-offs previously proposed in literature for each specific assay. Interestingly, in our population, a high percentage of women with polycystic ovarian morphology (PCOM) showed AMH levels lower than the cut-offs proposed. Serum AMH levels were significantly higher in patients with phenotype A compared to all other phenotypes for all three assays. For all three assays, we found a statistically significant positive correlation between AMH levels and LH, LH/FSH ratio, testosterone, androstenedione, FAI, mean follicular number, and mean ovarian volume. In contrast, AMH levels were negatively correlated with BMI in all groups. Limitations, reasons for caution This is a retrospective cross-sectional study. Longitudinal studies with a large cohort of women that could be followed for many years are needed to validate our nomogram. Wider implications of the findings AMH threshold levels can be influenced by several variables, making it difficult to identify a general cut-off to diagnose PCOM. Further research should overcome the concept of a single cut-off value. The results of the present study may help clinicians interpret AMH levels in PCOS patients across age ranges. Trial registration number In this retrospective cross-sectional study, 2,725 women who consulted the Reproductive Endocrinology and Infertility Clinic of the Erasmus University Medical Center, Rotterdam, The Netherlands, were included. The Medical Ethical Review Board of the Erasmus University Medical Center Rotterdam approved retrospective studies within this patient population, which includes women with ovulatory dysfunction (MEC-2020-0534).