Abstract: 【Objective】 This study aims to develop a green, low-cost, rapid detection technique for inulin and total sugar content in the roots of Taraxacum kok-saghyz Rodin(TKS) using near-infrared (NIR) spectroscopy, thereby advancing research in cultivation and breeding of TKS. 【Methods】 This study involved 103 dry root powder samples of TKS (Rubber Grass). Spectra were collected using the FOSS NIRSTM DS2500F SR near-infrared spectrometer (wavelength range 850-2500 nm) by Foss Company. Enzyme labeling method was employed to measure the total sugar and reducing sugar content in the root samples and the inulin content was calculated by subtracting the reducing sugar content from the total sugar content. The SPXY algorithm was utilized to divide the samples into training and validation sets. Pretreatments on the raw spectral data included Moving Window Smoothing (MWS), Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), and First Derivative (FD). Competitive Adaptive Reweighted Sampling (CARS) was applied for wavelength selection, followed by establishing Partial Least Squares (PLS) prediction models for inulin and total sugar content in TKS roots. 【Results】 The content range of total sugar, reducing sugar, and inulin in the sample set was 9.79%-51.85%, 3.11%-8.98%, and 6.41%-45.35%, respectively, with variation coefficients of 34.18%, 32.61%, and 39.46%. The samples were divided into calibration and validation sets at a ratio of 4∶1 using the SPXY algorithm, with each set containing 82 and 21 samples for both total sugar and inulin. The optimal PLS prediction model for inulin content utilized MWS-SNV-CARS preprocessing, achieving a validation set correlation coefficient (Rv) of 0.942, a root mean square error (RMSEv) of 2.515, and a relative performance deviation (RPDv) of 2.977. For total sugar, the best PLS model also employed MWS-SNV-CARS preprocessing, reaching an Rv of 0.949, RMSEv of 2.490, and RPDv of 3.175. Both models exhibited RPDv values exceeding 2.5, indicating good performance. 【Conclusion】 NIR spectroscopy proves to be an effective method for the green, low-cost, and rapid quantitative determination of inulin and total sugar contents in the roots of TKS.