For ambulatory children with cerebral palsy, the assessment of walking energy efficiency is utilized to determine functional changes following surgical, pharmacologic, or orthotic interventions. While the assessment of energy efficiency is considered a useful outcome tool, minimal information exists about the changes in energy efficiency over one year in children with cerebral palsy at different gross motor function classification system (GMFCS) levels and whether the patterns of change are similar to their able-bodied peers. The purpose of this study was to determine whether energy efficiency variables change similarly over one year in children with cerebral palsy by GMFCS level and whether they differ from their age-matched peers. Forty-five able-bodied children and 34 children with cerebral palsy, GMFCS levels I-III participated in the study. Energy efficiency variables were measured at baseline and at 12 months using a Cosmed K4b2. All subjects walked at their self-selected velocity for testing around a 33 m track. Baseline velocity and net non-dimensional cost (NNcost) differed by GMFCS level and between the able-bodied peers and all GMFCS levels. Children in GMFCS level III had the highest cost and the slowest velocity. When controlling for age and baseline values, significant differences in the magnitude of change were seen in velocity between children in GMFCS level III and children in GMFCS level I and II and their able-bodied peers. In comparison to their able-bodied peers, all GMFCS levels had an increase in NNcost over one year when controlling for age and baseline NNcost, with the difference in magnitude increasing by GMFCS level. Consistent with the literature, children with cerebral palsy had an increase in NNcost over one year in comparison to their able-bodied peers, which increased with GMFCS level. This finding demonstrates that when evaluating the change in walking energy efficiency with maturation and therapeutic intervention, comparisons should be made by GMFCS level.