The work aims to explore the mathematical prediction model for the lead release in ceramic glaze and establish a rapid extraction and detection method for toxic metal lead. The data of release of lead extracted at different temperatures and at different time were measured by the extraction experiment. Then, the relationship among lead release, dissolution time and dissolution temperature was analyzed to find the law of lead release. A mathematical prediction model of the lead release versus time and temperature was established by fitting the experimental data, thus achieving the rapid extraction and determination of the lead. The established lead release prediction model fitted the release of lead extracted by the acetic acid with volume fraction of 4%, and the correlation coefficient (R2) was greater than 0.98. The proposed model was verified by the release of lead in acetic acid solution with volume fraction of 8% and 14%, and the results showed that R2 was greater than 0.98. The model parameters fitted with the experimental data for the first 5 h were used to predict the lead release for 8-24 h. Pearson correlation coefficient between the predicted results and the experimental data was greater than 0.99. The lead release prediction model can properly fit the experimental data and has strong universality. It can predict the lead release for 24 h at 22 ℃, so as to achieve rapid extraction and determination of the lead in ceramic glaze.