In order to estimate the ability of biochar to sequester carbon as part of greenhouse gas removal technology, there is a need for rapid and accessible estimations of biochar stability. This study employs a novel method using Fourier transform infrared spectroscopy (FTIR) to predict common stability indicators, namely H:C and O:C molar ratios. Biochars derived from barley straw were produced at temperatures from 150 to 700 °C. The greatest compositional changes of the biochars occurred between 200 and 400 °C. All biochars produced at ≥400 °C achieved H:C < 0.7 and O:C < 0.4, indicative of biochars suitable for soil application. Regression models were built using FTIR data to predict H:C and O:C molar ratios. The H:C model produced a coefficient of determination (R2) of 0.99, mean absolute percentage error (MAPE) 6.86%, and root-mean-square error (RMSE) of 0.07. The O:C model achieved the same R2 (0.99), MAPE of 9.02%, and RMSE of 0.03. Our results demonstrate that combining FTIR data with modeling is a promising rapid and accessible method for attaining biochar stability data.