<p>This study presents the electrochemical micromachining of a magnesium alloy using an in-built micromachining setup with a minimum quantity electrolyte arrangement (MQE). A steel tool with a diameter of 600&#xa0;µm and sodium nitrate as the electrolyte was used to machine microholes in workpieces of size 20 × 20 × 2&#xa0;mm. The machining performance of the second method, MQE, was compared to the first method. The main outputs that were measured were material removal rate (MRR) and overcut (OC). Material Removal Rate (MRR) is a measure of machining characteristics that was taken based on depth-based approximation, implying constant tool geometry. The amount of material removal was taken by optical microscopy-based measurement of microhole diameter. Using important input parameters like machining voltage, duty cycle, and electrolyte concentration, MRR and OC are modelled and predicted with the help of machine learning algorithms like Gradient Boosting Regression and Support Vector Regression (SVR). When used together, these algorithms can model complex nonlinear relationships that govern machining results with exceptional accuracy and enhance prediction and process analysis. The results show that the proposed method can accurately predict machining performance and is useful for selecting parameters of electrochemical micromachining.</p> Graphical Abstract <p></p>