The traditional methods of property valuation, typically relying on market comparable and expert judgment, often lead to inaccurate pricing, which affects market stability and investor confidence. In developed countries, Artificial intelligence techniques have increasingly been adopted to enhance the accuracy of property price predictions, addressing issues of overpricing and underpricing. However, in developing countries like Nigeria, the adoption of these advanced methods remains limited. This study aims to bridge this gap by evaluating the accuracy of four AI techniques in predicting residential property prices in the Lagos Metropolitan area. The selected AI techniques including Random Forest, Bagging Regressor, Artificial Neural Network and Extra Tree Regressor. A total of 3,079 datasets utilized in this study were extracted from the databases of 53 estate surveying and valuation firms licensed to assess the value of land and buildings within the Lagos Metropolitan residential property market. These datasets underwent random partitioning, with 80% allocated for training purposes and the remaining 20% designated for testing. Performance metrics, including computational time, Root Mean Square Error (RMSE), Coefficient of Determination (R2), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) were employed to assess the predictive accuracy of the models under review. The findings indicate that all four models effectively predicted residential property prices within the study area. Notably, the Extra Tree Regressors exhibited superior performance in terms of both consistency and stability in prediction, while the Bagging Regressor emerged as the fastest computational technique among those examined. This paper emphasizes the importance of selecting techniques based on task-specific criteria rather than relying solely on general accuracy. While all four models successfully captured the overall trend in property prices, disparities between predicted and actual values suggest room for improvement. Operations such as crossvalidation, hyperparameter tuning, and the inclusion of additional price predictor variables are identified as potential avenues for enhancing the predictive accuracy of the selected models. The findings of the study offer valuable insights for real estate professionals, investors, policymakers, and other stakeholders in the field.