This paper evaluates the performance of the on-the-fly direct dynamics variational multi-configuration Gaussian (DD-vMCG) method using three molecular systems proposed by Ibele and Curchod as molecular versions of the Tully model systems. The Ibele–Curchod models, named IC1 (ethylene), IC2 (4-N,N-Dimethylaminobenzonitrile, DMABN), and IC3 (fulvene), are chosen for their different deactivation pathways after excitation to their ππ* bright states. The study compares the DD-vMCG method with the Tully Surface Hopping (TSH) method and other non-adiabatic dynamics methods, such as ab initio multiple spawning (AIMS) and multi-configurational Ehrenfest (MCE). The potential surfaces for each system are also tested and compared to on-the-fly results. The results highlight the differences in population dynamics between the methods, particularly in the case of IC1, where the TSH and vMCG simulations show significant discrepancies. The study emphasizes the importance of a common set of test systems and a mutually accepted approach to represent dynamical results for a fair comparison of different methods.This paper evaluates the performance of the on-the-fly direct dynamics variational multi-configuration Gaussian (DD-vMCG) method using three molecular systems proposed by Ibele and Curchod as molecular versions of the Tully model systems. The Ibele–Curchod models, named IC1 (ethylene), IC2 (4-N,N-Dimethylaminobenzonitrile, DMABN), and IC3 (fulvene), are chosen for their different deactivation pathways after excitation to their ππ* bright states. The study compares the DD-vMCG method with the Tully Surface Hopping (TSH) method and other non-adiabatic dynamics methods, such as ab initio multiple spawning (AIMS) and multi-configurational Ehrenfest (MCE). The potential surfaces for each system are also tested and compared to on-the-fly results. The results highlight the differences in population dynamics between the methods, particularly in the case of IC1, where the TSH and vMCG simulations show significant discrepancies. The study emphasizes the importance of a common set of test systems and a mutually accepted approach to represent dynamical results for a fair comparison of different methods.