The next generation of optical telescopes are expected to see first light at the dawn of the next decade. Out of the 3 ongoing ELT projects only the GMT has selected an optical high-resolution spectrograph as a first light instrument. The G-CLEF echelle spectrograph, being built by a consortium led by Harvard-Smithsonian Ctr for Astrophysics, has been designed to serve a broad array of science goals & projects, such as detection & characterization of Earth-like planets & measuring chemical abundances in stars in the Milky Way & dwarf satellite galaxies & globular clusters. Multiplexing capabilities are also being considered. After a short overview, I will describe the essential design aspects of the G-CLEF, focusing on the challenge of building an instrument that simultaneously fulfills the need of precision radial velocity measurements, good blue efficiency for stellar abundance work & also being capable of multi-object spectroscopy of very faint stars. We have developed a design that includes high internal efficiency, several resolution modes & a spectral coverage from 3500-9000 A & a multi-object capability of up to ~30 targets. I’ll describe our efforts towards enabling radial velocities precision measurements at the 10 cm/s level.