If the baryonic content of galaxies consists primarily of stars, ISM,  and hot (10^7 K) x-ray halo gas, then galaxies are missing between 70 - 95 % of their baryons relative to the cosmological fraction. When accounting for the baryon budget of galaxies, however,  we must not overlook the cooler (10^4 K) photo-ionized gas phase that makes up the circumgalactic medium (CGM). Our collaboration, COS-Halos, has been working to characterize the elusive multiphase CGM that extends out to at least 300 kpc from stellar components of galaxies. Specifically, we have observed the halo gas of 50  galaxies drawn from the imaging dataset of the Sloan Digital Sky Survey (SDSS) whose angular offsets from quasar sightlines and redshifts imply impact parameters (rho < 150 kpc) well inside their virial radii.  As we have shown in previous empirical studies, these data comprise a carefully-selected statistically-sampled map of the physical state and metallicity of the CGM for L ~ L* galaxies. Of particular relevance to the halo missing baryon problem is the total baryonic mass contained in the multiphase CGM, as traced by absorption from hydrogen and metal lines in various ionization states (e.g. MgII, SiII, CII, SiIII, CIII, SiIV, OVI).  In this talk, I will describe how I have modeled the photoionized gas of the CGM with a range of physical conditions, and rigorously determined the CGM gas ionization parameters and metallicities along 33 of the COS-Halos sightlines that provide the best-determined measurements of HI and metal-line column densities.  With the constraints imposed by the data and models, I am able to provide the most reliable mass estimate of the CGM to date, and show definitively that the CGM is an important reservoir of baryons on galactic scales.