Paeonol shows a wide range of pharmacological activities in the pharmaceutical industry. In this study, the solubility of paeonol in supercritical carbon dioxide (ScCO2) is determined by measuring the cloud-point pressures of the binary system in a high-pressure variable-volume view cell at various temperatures (i.e., 313.2, 318.2, 323.2, 328.2, 333.2 K) and pressure range of 8-15 MPa. The results show that paeonol has good solubility in ScCO2, and the solubility increases obviously with decreasing temperature and increasing pressure in the experimental range. Then, four density-based models, namely, Bartle, Mendez-Santiago and Teja, Ch and Madras, and Gordillo, are employed to correlate and predict the solubility data. The relevant average absolute relative deviation values of Bartle, Mendez-Santiago and Teja, Ch and Madras, and Gordillo are 9.05, 9.37, 8.75, and 8.49%, respectively, indicating that the models can well predict the solubility data.