The AgMIP project has the goal of using multiple crop models to evaluate climate impact on agricultural production and food security in developed and developing countries.  There are several major limitations to this goal, including the need to cross-train modelers in crop models other than their normal favored model, plus the inability to use the same input files for running simulations with both models.  Two activities were followed to address these shortcomings among regional teams that desired to use multiple models to evaluate climate impacts on food security and economic consequences.  Therefore we designed and conducted courses in which participants trained on two different sets of crop models, with emphasis on the alternate crop models.  In a second activity the IT group within AgMIP created templates for inputting soils, management, and crop data into AgMIP data bases, and developed translation tools for converting the entered data from the data base or DSSAT files into files ready to run for other crop models such as APSIM or STICS.  The strategies for conducting this multi-model course and developing entry and translation tools will be discussed.  Participants came to the course with data already in-hand and being simulated with their favorite model.  Following lectures on the general principles of crop growth and soil water balance common to both models, the participants used the IT Tools to convert the files to the alternate model, and worked with the experts of those models to calibrate the new models for the cultivars and then to simulate production associated with farmer field survey data.