This paper investigates the effectiveness of four machine learning models namely Linear Regression (LR), Decision Tree Regression (DTR), K-Nearest Neighbors Regression (KNNR), and Random Forest Regression (RFR) in forecasting current, voltage, and temperature of a 10 A h Li-ion 3S4P battery module during charging cycles. The models were trained and tested on data collected at various C-rates (0.5C, 1C) and ambient temperatures (30, 35 °C). The MSE and R2 metrics were used to compare and publish the performance findings. Results show that RFR and KNNR models outperformed other models, with KNNR model demonstrating greater forecasting capabilities for intermediate unseen data (0.75C). This research highlights the potential of machine learning models in accurately predicting Li-ion battery behavior, enabling improved battery thermal management and optimization.