Understanding the atmospheric and evolutive properties of very low mass stars, brown dwarfs, and gas giant exoplanets have been important challenges for modelers around the world since the discovery of the first brown dwarfs in the Pleiades cluster (Rebolo et al. 1995) and in the field (Nakajima et al. 1995). The early studies of brown dwarfs have provided rich insights into atmospheric physics, with discoveries ranging from cloud formation (Tsuji et al. 1996), methane bands (Oppenheimer et al. 1995) and ammonia bands (Delorme et al. 2008), to the formation of wasi-molecular KI-H2 absorption (Allard et al. 2007), and to disequilibrium chemistry (Yelle & Griffith 2001). New classical 1D models yield spectral energy distribution (SED) that match relatively well despite these complexities. These models have for instance explained the spectral transition from M to L, T and now Y brown dwarf spectral types (Allard et al. 2013). However, in presence of surface inhomogeneities revealed recently for a nearby (2 pc) brown dwarf (Crossfield et al. 2014), the SED may well fit even exactly, but the model parameters could be far from exact, e.g. with the effective temperature by several hundred kelvins too cool in the case of dusty brown dwarfs and young gas giant exoplanets! I will review the progress achieved in reproducing the spectral properties of very low mass stars, brown dwarfs and gas giant exoplanets, and review progress in modeling more accurately their atmospheres using Radiation HydroDynamical (RHD) simulations.