The study was to accurately evaluate the quality of table grape,18 different varieties of table grapes were selected. 14 basic quality indicators were measured and analyzed by sensory scores and correlation analysis. Principal component analysis was employed to build a composite score mathematical model. The results showed that there was a significant difference in the quality indexes among different grape varieties, among them, hardness difference was largest (coefficient of variation is 51.18%). Correlation analysis showed that there was an independence and a certain degree of correlation between indicators. Principal component analysis was extracted from 4 principal component factors, and cumulative contribution rate reached 88.63%. The comprehensive score comfirmed that the best grape was Black Grape, Shine-Musca, Wink and Jumeigui; The quality of Kyoho Grape, Ruby Seedless and Gold finger were relatively poor.