Background: Triamterene is a potassium-conserving diuretic mainly used in the treatment of edematous diseases. Although there are some methods to measure the content of triamterene, the existing methods have the characteristics of sample destructiveness and low flux, making it difficult to meet the needs of online monitoring. In recent years, transmission Raman spectroscopy (TRS), as a new technology for the determination of drug content, has emerged as being nondestructive and rapid and provides a new method for the determination of triamterene content. Objective: In this study, we used transmission Raman spectroscopy combined with partial least squares (PLS) approaches to establish a six concentration levels model for measuring the content of triamterene. The model was applied to determine whether tablets are commercially available. Methods: Firstly, TRS was used to collect the spectra of the principal components and mixed excipients of triamterene, and the feasibility test was carried out. Secondly, six concentration levels were determined by the design of experimental (DOE), and hand-made tablets were prepared to obtain corresponding spectra. The content prediction model was established by the PLS method, and the content of triamterene was determined by high-performance liquid chromatography (HPLC) to correct the model. Finally, the model was applied to the determination of triamterene in commercially available tablets. Results: The results showed that the established model was successfully applied to the determination of triamterene. The values of RMSEC (0.0089783) and RMSECV (0.0097241) of the final model were very low and close to each other. The relative error of 12 hand-made tablets predicted by this model was less than 5% compared with the results determined by HPLC. In addition, in the process of applying the model to the determination of the content of commercially available tablets, the accuracy of the model can be significantly improved by adding the spectrum of commercially available tablets. The corrected model was used to determine the content of triamterene in two commercially available tablets. The results showed that the relative error was less than 5%. Conclusion: We described a new strategy to analyze the content of active pharmaceutical ingredients (API) in triamterene tablets by TRS with PLS. The established model has the advantage of non-destructive and rapid quantitation, which can provide a new method for real-time monitoring of production lines and promote the development of process analytical technology (PAT).