In this work, feed-forward back-propagation Artificial Neural Networks (ANNs) have been presented to predict the enhancement of the relative thermal conductivity and viscosity of a wide range of nanofluids with different base fluids and nanoparticles. The thermal conductivity ratio of nanofluids with respect to the base fluids has been modeled using an ANN model. The model considers the effects of the thermal conductivity of the base fluid, the thermal conductivityof nanoparticles, nanoparticle volume fraction percent, temperature, and nanoparticle cluster average size. The total number of experimental data used to design the stated network is 483 from 18 different nanofluids. The (5-18-1) topology has been obtained as the best topology of the ANN model. The results of the AARD% for the train, validation, and test sets of data are 2. 6, 2. 2, and 2. 3, respectively. The viscosity ratio of the nanofluids with respect to the base fluids has been modeled using the other ANN model. The viscosity of the base fluid, density ratio of the base fluids with respect to the nanoparticle, nanoparticle volume fraction percent, temperature, and nanoparticle cluster average size have been selected as the inputs of ANN model. The 510 experimental data have been used to design the stated network. The (5-19-1) topology has been obtained as the best topology of the ANN model. The results of the AARD% for the train, validation, and test sets of data are 2. 9, 3. 1, and 3. 2, respectively. Accordingly, two studied ANN models are in good agreement with experimental data. A comparison between the predictions of the proposed ANN models and those predicted by some traditional models such as Maxwell and Bruggeman models shows that much better agreements can be obtained using the ANN model. This model also can able us to predict the relative thermal conductivity and viscosity of new nanofluids in different conditions.