Climate factors have distinct impacts on crop yields. Understanding the relative importance of these factors to crop yield variation could provide valuable information about crop planting and management under climate change condition for policymakers and farmers. The current study investigated the applicability of support vector machines (SVMs) in determining the relative importance of climate factors (mean temperature, rainfall, relative humidity, sunshine hours, daily temperature range, and rainy days) to yield variation of paddy rice in southwestern China. The SVM models were compared with traditional artificial neural networks and multiple linear regression. The performance accuracy was evaluated by mean absolute error (MAE), mean relative absolute error (MRAE), root mean square error (RMSE), relative root mean square error (RRMSE), and coefficient of determination (R 2 ). Comparative results showed that SVMs outperformed artificial neural networks and multiple linear regression. The SVM with radial basis function performed best with MAE of 0.06 t ha −1 , MRAE of 0.9 %, RMSE of 0.15 t ha −1 , RRMSE of 2.23 %, and R 2 of 0.94. The results showed that SVMs are suitable for determining the effects of climate on crop yield variability. The relative importance of the studied climate factors to rice yield variation was in order of sunshine hours > daily temperature range > rainfall > relative humidity > mean temperature > rainy days in the current study area.