We have applied the InfraRed Flux Method (IRFM) to a sample of 475 dwarfs and subdwarfs in order to derive their effective temperatures with a mean accuracy of about 1.5%. We have used the new homogeneous grid of theoretical model atmosphere flux distributions developed by Kurucz (1991, 1993) for the application of the IRFM. The atmospheric parameters of the stars cover, roughly, the ranges: 3500K<=T_eff_<=8000K -3.5<=[Fe/H]<=+0.5 3.5<=log(g)<=5. The monocromatic infrared fluxes at the continuum, and the bolometric fluxes are derived using recent results, which satisfy the accuracy requeriments of the work. Photometric calibrations have been revised and applied to estimate metallicities, although direct spectroscopic determinations were preferred when available. The adopted infrared absolute flux calibration, based on direct optical measurements of angular stellar diameters, sets the effective temperatures determined using the IRFM on the same scale than those obtained by direct methods. We derive three temperatures, T_J_, T_H_ and T_K_, for each star using the monochromatic fluxes at different infrared wavelengths in the photometric bands J, H, and K. They show good consistency over 4000 K, and no trend with wavelength may be appreciated. We provide a detailed description of the steps followed for the application of the IRFM, as well as the sources of the errors associated to the different inputs of the method, and their transmission into the final temperatures. We also provide comparison with previous works.