The main objective of this study was to implement and evaluate a computer-based nutrition education intervention in the primary schools of Pakistan and to explore the acceptability of such a program in the Pakistani context. The study design was multifactorial with repeated measures of nutrition knowledge, taken at three points in time (baseline, post-intervention and follow-up) from intervention and comparison groups. Qualitative methods (focus group and observations of lessons) were also employed. Three hundred and forty four children (8-10 years old) of eleven primary schools of Lahore participated in the study. In a two week school-based nutrition education intervention, comparison group received nutrition education through worksheets, card and board games. Intervention group used computer-based tool along with work-sheets card and board games. Control group was not provided nutrition education. Nutrition knowledge was measured in each child at baseline, at the end of two-week intervention, and at three - months' follow-up with a validated nutrition knowledge questionnaire. The nutrition knowledge increased significantly in both intervention and comparison groups (P<0.001). No tangible difference was observed at post-intervention and follow-up between the two groups. The intervention effect was significantly maintained at three months' follow-up, indicating the retention of the nutrition knowledge. Long term knowledge gain and retention of knowledge is significantly higher among girls as compared to boys. Younger children (8-9 years) significantly (p=0.037) retained more nutrition knowledge at three months follow-up. Computer-based tools were found helpful in disseminating nutrition education. However, they were not the panacea. An amalgam of computer-based tools and other creative methods brings more fruitful results than using these methods in isolation. The low level of nutrition knowledge of Pakistani children urgently demands incorporation of the nutrition education in the school curriculum.