To accurately detect the content of major metal elements in agribiochar using laser induced breakdown spectroscopy (LIBS) and improve its poor detection sensitivity, high temperature treatment was proposed to remove the effects of moisture, fixed carbon and organic matrix. Primarily, totally 66 representative agribiochar samples with Aad content ranging from 28% to 42% were collected and divided using multiple Aadpartition intervals (14%, 7%, 3.5% and 2%). Moreover, when the interval value was set to be 7%, the Aadpartition of the collected samples was divided into 28%~35% (38 samples) and 35%~42% (28 samples). And the corresponding determinant coefficient between raw samples and treated samples was higher than 0.96. Thus, it was possible to develop a serial Aadpartition calibration model using spectral information of treated samples and chemical information of raw samples. In comparison with the modeling effects of raw samples, the partial least squares regression (PLSR) models developed by treated samples had lower values of relative standard deviation of crossvalidation set. The pairwise T test of its prediction set showed that there was no significant difference between the measurement of LIBS and inductively coupled plasma mass spectrometry (ICP-MS). The results showed that the LIBS can be used to simultaneous, accurate and quantitative analysis of major metal elements in agribiochar based on the high temperature treatment and serial Aadpartition calibration model.