One of the main ingredients of current stellar population models is a library of stellar spectra. Both empirical and theoretical libraries are used for this purpose, and the question about which one to use is still being debated in the literature. Empirical and theoretical libraries are improving significantly over the years, and many libraries have become available lately. It is not clear what are the advantages of using each of these new libraries, and how far behind are models compared to observations. Here we compare in detail some of the modern theoretical libraries availabe in the literature against empirical libraries attempting to detect their weaknesses and strengths. The aim is to be able to compute in the short future a new synthetic stellar library that combines the best qualities of the current available ones, while improving considerably their weaknesses.
