For the landslide susceptibility prediction (LSP) based on machine learning (ML) models, the reasonable selection of negative samples has an important influence on the LSP performance. Generally, the main selection methods include randomly selecting from the whole study area or from the specific attribute areas such as low slopes. The negative samples selected by the above methods are often inaccurate or biased, resulting in low accuracy and low reliability of LSP. To solve this problem, the coupling model of ML and information value (IV) method was proposed for LSP. Taking Ruijin City as the study area, the attribute values of the environmental factors were transformed into the IV values of the contribution to the landslide to obtain the very low and low susceptibility areas. The negative samples were randomly selected in the above areas for the training and validation of machine learning models. The new coupling models of IV–SVM and IV–RF were constructed for the LSP of Ruijin. Further, IV–SVM and IV–RF models were compared with the single SVM and RF model with negative samples randomly selected from the whole study area, as well as the low-slope SVM and RF model with negative samples randomly selected from specific attribute areas with a slope less than 2°. Finally, Kappa coefficient (KC) and receiver operating characteristic (ROC) curve were used to verify and compare the modeling results. The AUC values of the ROC curve and KC of IV–SVM and IV–RF models were 0.828, 0.920 and 0.876, 0.988, which were higher than those of single SVM, RF model and low-slope SVM, RF model, respectively. Meanwhile, IV–SVM and IV–RF models have a smaller mean value and larger standard deviation of a susceptibility probability distribution. Results showed that: 1) IV–SVM and IV–RF models had the higher LSP accuracies than those of the single SVM, RF model and low-slope SVM, RF model, respectively; 2) RF model had higher LSP accuracy compared to the SVM model; 3) The coupling model such as IV–RF could address the inaccuracy of negative sample sampling existing in the single model and the shortcomings of the low slope model in the selection of slope interval, thus improving the LSP accuracy. In conclusion, this study provided a new idea for the negative sample sampling method for LSP using ML models