The thrust force has a direct effect on important output responses such as surface roughness, power consumption, and tool wear. The measurement process for thrust forces is very time-consuming and costly. Low success rates are achieved in the estimation of thrust force. The mechanical properties and chemical composition of the workpiece material were not taken into account as input properties in the previously established models. However, these features are of high importance in the formation of the thrust force. In this study, it is aimed at estimating the thrust force with the least input feature with a high success rate by establishing single and hybrid models with different machine learning algorithms. As a result, the thrust force can be successfully predicted by machine learning algorithms by entering two properties. Successful thrust force estimation will provide more accurate process planning, material selection, and optimization.