Yes use an ensemble of the better models that are available, but don't use all variables. Choose those that have biological reasoning for your organism. Average temperatures and precipitations mean little, so use those that describe extreme events that limit your organism. Average temperature of the warmest period, minimum temperature of the coldest period, temperature variability, precipitation variability, precipitation of the wettest and driest periods. Avoid composite variables where they are based on the precipitation of the warmest or coldest period or temperature of the wettest or driest period as these datasets are internally flawed.