Abstract: To evaluate the transfer effects of near-infrared (NIR) spectral quantitative models for the main conventional chemical indexes of tobacco using polymers as standard samples, and to promote the application of NIR spectroscopy in the tobacco industry, spectral transfer models were established for the same brand and different brands of NIR spectrometers from slave to the same host instruments using piecewise direct standardization (PDS) and spectral space transformation (SST), with tobacco and 17 polymers as the standard materials. Principal component analysis and Mahalanobis distances between the transferred and original spectra from the host instrument were used to optimize the PDS and SST parameters. The effectiveness of the transfer models was evaluated using the root mean square error of prediction (RMSEP) between the predicted values based on the transferred spectra from the slave instrument and the determined values obtained using reference methods. The results showed that: 1) The optimal PDS parameters were a segment wavelength window width of 6 cm^-1 and six PLS principal factors. 2) The optimal SST parameter was six PLS principal factors. 3) The RMSEP values for total nitrogen, nicotine, and total sugar in the 90 external test samples across the same brand and different brand NIR spectrometer model transfers indicated that the calibration performance of the spectral transfer models was as follows, ranked from best to worst: tobacco standard sample + SST, tobacco standard sample + PDS, polymer standard sample + PDS, and polymer standard sample + SST. Tobacco performed better than the polymers as standard samples for transferring NIR models of total nitrogen, nicotine, and total sugar in flue-cured tobacco leaves. These findings provide a foundation for the wider application of NIR spectroscopy in the tobacco industry.