The rheology of drilling mud is significantly affected by viscosity, salt concentration, temperature, and nanoparticle concentration. This study uses two machine learning techniques to predict the viscosity and filtration loss of water-based mud containing zirconium dioxide (ZrO2) nanoparticles as a function of concentration, shear rate, temperature, time, and differential pressure. The techniques utilized are artificial neural network (ANN) and random forest (RF). Both machine learning algorithms are tuned to acquire the best set of hyper-parameters. Results yielded by the algorithms are compared using statistical error matrices. The predicted results achieve R2 value higher than 0.9 for both models. The comparative analysis from the outcomes of both models exhibits that shear rate and time contributed the most to the variation in viscosity and fluid loss, respectively. After further validation of developed models with experimental data, a good agreement between predicted and experimental data is found.