Upgrading raw biogas and landfill gas to methane purity [98 % is a vital prerequisite for the utilization of biogas for compressed natural gas pipeline applications. Pressure swing adsorption (PSA) is an industrial separation technology widely used for the separation and purification of methane rich streams from raw biogas and landfill gas. Current PSA technologies make use of differential adsorption of gases on traditional adsorbents, such as zeolites, activated alumina and molecular sieves. In order to evaluate the potential of new adsorbent materials, such as metal-organic frameworks (MOFs) for PSA applications, and for PSA process development, thermodynamic equilibrium adsorption isotherms data of the pure components and mixtures and their adsorption isosteric heats are required. In this work we use extended Langmuir (ELM) model and Doong-Yang multi-component (DYM) adsorption model to predict the isotherms of biogas compositions containing binary and ternary mixtures of CO 2 , CH 4 and N 2 on activated carbon Maxsorb and metal-organic framework Cu-BTC. The model parameters required for predicting the mixture adsorption isotherms using the ELM and DYM are obtained, respectively from the singlesite Langmuir and Dubinin-Astakhov non-linear regression of pure gas isotherms experimentally measured at 298 K and over a pressure range of 0-5 MPa. Predicted data are compared with the experimental binary and ternary mixture adsorption isotherms on Norit R1 extra and Cu-BTC available in the literature. Selectivity and thermodynamic delta-loading of equimolar mixtures of CH 4 and CO 2 on Cu-BTC and Maxsorb are determined from the predicted mixture isotherms and are compared with that of a traditional PSA adsorbent, zeolite 13X.