Abstract The high temperature heat extraction from a porous media reciprocal flow burner (RFB) is studied numerically. In RFB, the air/fuel mixture flow direction is switched periodically to contain and sustain combustion. High temperature heat extraction is essential for the efficient operation of small electrical and thermal systems based on thermoelectric generators, Stirling engines and micro-turbines. In this paper, the efficiency of the heat extraction is studied for different extraction locations, temperatures and equivalence ratios. An 1-D time-averaged model is developed and validated to represent a RFB and to further predict the behavior of the burner. Optimization on the burner’s geometry is also performed to increase the heat extraction efficiency.