Taxus baccata is a relic-declining species, especially at the southern edge of its natural distribution. On the Island of Sardinia, we found 237 locations where yews grow, being indifferent to substrates, preferring moist sites with northern aspects and with an elevation range of 695–1,225 m a.s.l. We applied the Maximum Entropy algorithm (MaxEnt), using 202 field-based sites and 19 bioclimatic variables, to model the present potential distribution. Our model showed high levels of predictive performances with significant values of AUC (training, 0.962; test, 0.963). Annual precipitation, mean temperature of the warmest month and isothermality were the key drivers for the distribution of yew in Sardinia. The response curves showed that T. baccata prefers annual mean temperatures ranging from 8.65 to 12.55 °C and annual mean precipitation from 770 to 1,085 mm. Most potential suitable conditions were found in northern and central Sardinia. We also modelled future habitat suitability for T. baccata under two different scenarios (ssp245 and ssp585), highlighting a predicted drastic reduction, up to >95% in the worst scenario, of the habitat suitability for yews by the end of this century. This is related to climate change and poses increasing concerns about the future of yews in Sardinia.