<b><i>Aim:</i></b> To evaluate the use of baseline inhibin B (INHB) levels to differentiate the progressive form (PF) from the non-progressive form (NPF) of central precocious puberty (CPP). <b><i>Methods:</i></b> A total of 62 girls were enrolled, 31 with PF and 31 with NPF. Using receiver operating characteristic (ROC) curves, we analysed the diagnostic performance of INHB in addition to other diagnostic tools used to differentiate the 2 forms of CPP. <b><i>Results:</i></b> INHB levels were higher in PF versus NPF (29.1 vs. 13.1 pg/ml; p < 0.001). The ROC area under the curve (AUC) was greatest for luteinizing hormone [LH; 0.807, standard error (SE) 0.069], followed by INHB (0.800, SE 0.067), ovarian volume (OV; 0.782, SE 0.070) and uterine volume (0.723, SE 0.076). In ROCs relative to a combination of such parameters, the AUC was greater for LH + INHB (0.972, SE 0.010), followed by OV + LH (0.841, SE 0.084) and OV + INHB (0.836, SE 0.075). The combination of INHB and LH (with cut-offs of 20 pg/ml and 0.2 IU/l, respectively) results in 98% sensitivity and specificity. <b><i>Conclusion:</i></b> Our results suggest that the addition of basal INHB values to baseline LH levels provides a reliable method to identify PF. Further replication studies are needed to definitively prove or disprove the utility and advantages of INHB levels as part of the work-up of CPP.