The total electron content (TEC) of the ionosphere is a parameter of great importance for the systems which use transionospheric radiowaves and for the checking of the validity of ionospheric models. It is known that the highest TEC values in the world occur at the peaks of the equatorial anomaly (EA) regions at -15¿ either side from the magnetic equator. In this paper, TEC measurements obtained with Faraday technique at three stations placed near both peaks of the EA are used to check the validity of three ionospheric models to predict TEC in these regions. The TEC data of the southern peak of the EA were obtained at Tucum¿n (26.9 ¿S, 294.6 ¿E) in 1982, and those corresponding to the northern peak were obtained at Palehua (21.4 ¿N, 203.2 ¿E) in 1978 and Delhi (28.6 ¿N 77.2 ¿E) in 1968. Ionosonde data from Tucum¿n, Delhi, and Maui, (20.8 ¿N, 203.5 ¿E) are also used. Two widely used models are considered, namely: the international reference ionosphere and the semiempirical low-latitude ionospheric model. The third model considered is a Chapman layer with scale height equal to atomic oxygen scale height (CHOEA). In general, the models underestimate TEC during daylight hours at Tucum¿n suggesting that the real electron density profiles are larger in electron density magnitudes than those given by the models. For Palehua and Delhi the models's predictions are better than those for Tucum¿n, and a very good agreement among calculated and measured TEC values at Palehua for some months is observed. Taking into account the simplicity of the TEC calculus with CHOEA, this model would be an adequate alternative to predict TEC at stations placed near the northern crest of the EA, at least for the considered solar activity. This model allows also to calculate TEC at the southern crest of the EA with a similar degree of accuracy of those values obtained with the other models. ¿ American Geophysical Union 1994