To enable the continuous production of cocrystal-containing pharmaceutical tablets, guaranteeing the cocrystal content of the final pharmaceutical tablets in the solid state is critical. This study demonstrates the quantification of caffeine-glutaric acid cocrystals in model tablets using transmission low-frequency Raman spectroscopy. Although distinguishing between cocrystals and raw materials using conventional Raman spectroscopy is difficult, the use of low-frequency Raman spectroscopy enables the discrimination of cocrystals and raw materials. Low-frequency Raman spectra were analyzed by the partial least-squares method (PLS) to obtain the predicted contents in the model tablets. To evaluate the quantitative ability of this method, the root means square error of cross-validation (RMSECV) was determined by comparing the actual concentration and predicted content with a calibration curve. For cocrystal-containing tablets, the quantitative ability of the transmission mode (RMSECV = 2.06- 3.17) was 13.4-31.4% higher than that of the backscattering mode (RMSECV= 2.37- 3.91). The coexistence of raw crystalline materials did not affect the quantitative ability for cocrystals.