Accurate knowledge of glacier bed geometry and flow parameters is essential for glaciological modeling and can be inferred from surface data and theoretical principles. This study evaluates the performance of the Instructed Glacier Model (IGM) v2.2.1 in estimating glacier ice thickness using the protocol from Phase II of the Ice Thickness Models Intercomparison eXperiment (ITMIX II). Results show that IGM ranks among the best-performing models, producing largely unbiased ice-thickness fields. Compensating errors, where underestimations of ice thickness in some glacier regions are offset by overestimations in others, are generally small, indicating robust overall performance. Extreme deviations from observed thicknesses are less frequent in IGM than in the ITMIX II model ensemble. We also examine the ability of IGM to reproduce glacier dynamics by comparing modeled and observed surface velocities. While IGM is generally capable of reproducing the surface velocities well, our analysis reveals a trade-off between matching ice thickness and surface velocity, possibly due to inconsistencies or biases in the observational datasets. Generally, the overall good performance of IGM in reproducing glacier geometry and dynamics supports its potential for initializing glacier evolution simulations.