Objective To establish and evaluate a reliable and highly reproducible neonatal rat model of hyperbilirubinemia and to provide an experimental basis for research of kernicterus and related mechanism of neuroinjury. Methods Sixty 7-day old SD rats (28 male and 32 female) were used in this study. Three doses of phenylhydrazine hydrochloride (25, 50, and 75 mg/kg) were intraperitoneally injected respectively to the neonatal rats to establish models of hyperbilirubinemia induced by hemolysis. The control group was set up at the same time. 48 hours after the experimental treatment, the bilirubin in blood and brain tissue, neuron-specific enolase (NSE) of brain tissue, and hemoglobin were detected to evaluate the models. Results Compared with the control group, the bilirubin in the blood and brain tissue and the brain tissue NSE in the three experimental groups were significantly higher than that in the control group (P<0.05), while hemoglobin content was significantly lower than that in the control group (P<0.05). The bilirubin of blood and brain tissue and brain tissue NSE in the 50 mg/kg and 75 mg/kg dose phenylhydrazine hydrochloride groups were significantly higher than that of the 25 mg/kg dose group (P<0.05), while hemoglobin content was significantly lower than that of the 25 mg/kg dose group (P<0.05). There were no significant differences between the 50 mg and 75 mg dose groups (P>0.05). Conclusions Intraperitoneal injection of phenylhydrazine hydrochloride can be used to produce neonatal rat models of hyperbilirubinemia, mimicking the clinical features of this disease, and 50 mg/kg of phenylhydrazine hydrochloride is the best concentration. It is an ideal method to establish newborn rat models of hyperbilirubinemia.