Tadalafil is an illegal additive in antifatigue supplements. It is often misused in various plant dietary supplements (BDS), resulting in serious health risks. In this paper, terahertz spectroscopy combined with chemometrics is used to quantitatively analyze the content of tadalafil in nutritional and health products. The absorption coefficient spectrum of tadalafil in the range of 0.1-2.5THz was obtained, and an obvious characteristic absorption peak appeared at 1.7THz. To verify the accuracy of this characteristic absorption peak theoretically, tadalafil was simulated by density functional theory, and the calculated terahertz vibration spectrum matched well with the experimental spectrum. Then, the pure fatigue-based nutraceutical matrix and pure tadalafil were mixed in different proportions, and the terahertz absorption coefficient spectra of the mixtures were obtained. Finally, a quantitative analysis model of the tadalafil mixture was developed based on the support vector regression (SVR) algorithm, and the SVR model was optimized using particle swarm optimization (PSO) and genetic algorithm (GA), respectively. Compared with the SVR model, both PSO-SVR and GA-SVR enabled some improvement in their prediction accuracy, but the PSO-SVR model ran faster at 4.85s, whereas the GA-SVR model had a higher prediction accuracy with a prediction set correlation coefficient (R P) of 0.9996 and a root mean square error (RMSEP) of 0.011. In summary, this study used terahertz time-domain spectroscopy for the identification and quantification of tadalafil in health product matrices. This study provides a new solution for the nondestructive detection of illegally added tadalafil in antifatigue health products, which is pivotal to the quality control of the health product industry.