The mid-Pliocene warm period has been suggested as an analog to a future warmer climate (Jansen et al., 2007), due to its similarity to present day in terms of land-sea distribution as well as CO2. The Pliocene Model Intercomparison Project (PlioMIP), established by Haywood et al. (2010, 2011), is an internationally coordinated effort to identify strengths and weaknesses of climate models in simulating of mid-Pliocene climate. PlioMIP is currently in its second phase, due to advances in the understanding of palaeogeography, ocean, terrestrial and cryosphere of the Pliocene epoch. Boundary conditions have been updated from PlioMIP1 and experiments were prescribed accordingly for PlioMIP2. In this study, a detailed description of the different components of the utilized Community Earth System Models (COSMOS, version: COSMOS-landveg r2413, 2009) as well as experimental procedures prescribed for PlioMIP2 (Haywood et al., 2016) are provided. Furthermore, the procedures applied to transfer the Pliocene Research, Interpretation and Synoptic Mapping (PRISM) Project mid-Pliocene reconstruction into model forcing is documented. The mid-Pliocene, as simulated by COSMOS and the PlioMIP2 prescribed boundary conditions is warmer than the PI-control simulation by 2.15 K in the global mean, corresponding to -0.08 K colder than the PlioMIP1 simulation by the same model. Higher-latitude warming, which is supported by the proxy evidences is still underestimated in the model outputs. The state of the Bering Strait, which is unknown during the mid-Pliocene (Gladenkov et al., 2002) was opened for PlioMIP1 and closed for PlioMIP2 simulations. This study show that a closed Bering Strait leads to better agreement of COSMOS simulation with the reconstruction in North Atlantic region that is very warm during the MPWP. Data-model discrepancies are not fully resolved even with the closure of the Bering Strait, this study show that uncertainty in the concentration of atmospheric CO2 contributes to the disagreement in various Ocean regions. Furthermore, climate sensitivity is estimated for climate states with two different background geographies, and the results confirmed that the response of global temperature to the doubling of CO2 is dependent on the background climate.