A computational model is developed in order to analyze flow, temperature and species distributions inside a 350 MW utility boiler. It is assumed that identification of high temperature or high velocity zones will help in the prevention of failures in the boiler walls, superheaters, reheaters and economizers. For the analysis of these failures the modeling of the chemical and physical phenomena inside the boiler is important, because one of the known causes of tube failure is the non-uniform heating of the tubes, which strongly depends on the combustion gases flow and temperature distributions. The 3-D computational fluid dynamics (CFD) codes provide an effective tool for this type of calculations. CFD calculations were performed for the condition of 100% of total load for a 350 MW utility boiler using either pulverized coal or heavy oil as fuels. The CFD calculations adopt a 3D-formulation of the mean flow equations in combination with the standard high-Reynolds-number k-epsilon turbulence model and a probability density function to model fuel combustion. The tube banks are represented by a porous media model. Comparisons between calculations and key global parameters from the power plant show relatively good agreement. Velocity profiles show a very complex flow in the boiler, especially in the lower part of the boiler, where the injected streams form a cyclone at the center of the boiler.