This investigation explores the magnetohydrodynamic (MHD) free convective heat and mass transfer characteristics of a ternary nanofluid traversing an exponentially accelerated inclined plate within a porous medium. The theoretical framework integrates the complexities of internal heat generation/absorption and fluctuating wall temperatures. Analytical solutions were rigorously derived utilizing the Laplace transform technique, while a sophisticated Artificial Neural Network (ANN) was implemented to forecast and corroborate these mathematical outcomes. Heat Transfer (Nusselt Number) evaluated against the interplay of the Prandtl number, thermal radiation parameters, and temporal progression. Mass Transfer (Sherwood Number) analyzed as a function of magnetic permeability, the Schmidt number, and time. Thermal Enhancement findings indicate that an augmentation in the nanofluid volume fraction significantly bolsters thermal conductivity, thereby elevating the temperature profile. The proposed Levenberg-Marquardt Algorithm-based Backpropagation Artificial Neural Network (LMA BANN) demonstrated exceptional predictive fidelity. The model achieved a precision threshold exceeding 99.9% for the Nusselt number and near-perfect accuracy for the Sherwood number. These results are substantiated by negligible Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) values, coupled with correlation coefficients (R) nearing unity, signifying a robust alignment between the analytical and predicted datasets.