Model-based approaches e.g., physiologically based pharmacokinetic (PBPK) modeling and allometric scaling, are widely used for the prediction of pediatric pharmacokinetics (PK) for different drugs. In this study, we compared these two approaches and evaluate their predictions of the pediatric PK for Tyrosine kinase inhibitors (TKIs). TKIs are used for the treatment of different malignancies in adults and children and an effective and precise approach for translating the well-established adult PK profile into a pediatric PK profile is much needed. Pediatric PK studies are difficult due to ethical and practical reasons, and extrapolation of PK profile in children from adults can lead to optimizing their dose in the pediatric population for better clinical outcomes. Two TKIs were selected (Imatinib, and Pazopanib) to evaluate the performance of each model-based approach. Adult population PK (popPK) models were selected from literature for both drugs then allometric scaling was applied to develop the pediatric popPK models. While for the PBPK approach, adult models for both drugs were built using each drug-specific parameters and reported physicochemical properties. Then the adult PBPK models were translated into pediatric models by changing the physiological characteristics (e.g., age, weight) of the population to the specified age group. The predictions of the pediatric PK profile from both model-based approaches were evaluated against the available observed PK data using visual predictive check plots and they demonstrated a good fit for the observed pediatric PK data but the allometric scaling approach for Imatinib performed better in terms of certainty of pediatric PK predictions.