Peach occupies an important share in the worldwide fresh fruit market. Soluble solid Content (SSC) is an important parameter for measuring peach quality. It is one of the most important internal properties that influence the consumer purchasing decision on fresh fruit and best picking period. Therefore, a handheld yellow nectarine SSC non-destructive detection device was developed based on visible and near-infrared spectroscopy technology. The hardware system of the device was mainly composed of a micro-spectrometer, two halogen lamps, an OLED display, a microcontroller, and a self-designed driven circuit. The real-time analysis and control software of microcontroller was written in C language with the help of Keil 5 development tool. Combined with the spectrum acquisition program written by LabView, the spectra of fruit sample were collected by the developed device. Yellow nectarines planted in Pinggu area in Beijing were used as fruit samples to evaluate the detection performance of the developed device. First of all, the visible and near-infrared spectra of samples in calibration set were obtained in the range of 680-940 nm. After pre-treated by average smoothing (window size=5) and maximum normalization, the spectra were used to develop a partial least squares (PLS) model for SSC prediction. The calibration and cross validation results obtained by the developed model had fairly good correlation coefficients and low prediction errors. To evaluate the robustness and accuracy of the developed model, the model was then applied to assess the SSC of samples in prediction set. The correlation coefficients and the root mean square error were 0.947 and 0.728%, respectively. The detection time of single fruit should not exceed 2 s. In order to improve the accuracy and stability of the model, the samples in calibration set and prediction set were combined as a new calibration set to develop an SSC prediction model, followed by the validation using an independent validation set. The results indicated that the correlation coefficient and the root mean square error of SSC prediction for yellow nectarine were 0.906 and 0.732%, respectively. However, the direct use of a calibration model built by one instrument to another instrument was impracticable, therefore, a calibration transfer or model transfer methods were needed to share a calibration model between same kinds of instruments. Calibration transfer was a mathematical method that made the spectrum measured by different samples or instruments as consistent as possible. Piecewise direct standardization (PDS) is a multivariate full-spectrum standardization method, which works in a similar principle to the DS algorithm, has become the most wide-used calibration transfer method. It was used to transfer the calibration model developed by the master device to the slave device in this study. After the model was transmitted from the master device to the slave device, the SSC was successfully predicted by the slave device, with the correlation coefficient and the root mean square error for SSC prediction in validation set of 0.865 and 0.919%, respectively. In addition, the developed handheld SSC detection device could transmit SSC prediction value by Bluetooth to cell phone application written by Android language, which achieved the visualization of SSC values on Baidu map based on the predicted SSC values and the mobile phone's own positioning information. The overall results of this study showed that the developed handheld detection device can achieve accurate and non-destructive measurement of yellow nectarine SSC. With the help of the model transfer algorithm, the model can be shared and effectively transferred between different devices, avoiding repeated modelling and saving a lot of time and money. The developed device could meet the demand for rapid, non-destructive, and on-site detection of fruit internal quality. There are good prospects for wide application of the developed handheld detection device.