BACKGROUND: Attention-deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder in children, often accompanied by working memory deficits. Recently, exercise interventions have gained attention as a potential strategy to improve cognitive function in children with ADHD. However, the effects of different types of exercise on working memory remain unclear. This study aimed to assess the effects of various exercise interventions on working memory in children with ADHD using a network meta-analysis. METHODS: A comprehensive search was conducted in PubMed, Cochrane, Embase, and Web of Science databases for relevant studies. After screening according to the inclusion and exclusion criteria, a total of 17 eligible studies were identified for analysis. A network meta-analysis was performed to integrate data and evaluate the effects of cognitive-aerobic exercise, ball games, mind-body exercises, interactive games, and general aerobic exercise on working memory in children with ADHD. RESULTS: The results indicated significant differences in the effectiveness of various types of exercise interventions on working memory in children with ADHD. Cognitive-aerobic exercise showed the most significant effect (SMD = 0.72, 95% CI: 0.44-1.00), followed by ball games (SMD = 0.61, 95% CI: -0.12-1.35). Mind-body exercises and interactive games had moderate effects (SMD = 0.50 and 0.37, respectively), while general aerobic exercise showed relatively small effects (SMD = 0.40, 95% CI: 0.19-0.60). SUCRA analysis further confirmed the highest preference for cognitive-aerobic exercise in improving working memory. Meta-regression analysis showed that intervention frequency and total intervention duration significantly affected the effectiveness of cognitive-aerobic exercise, while other variables did not significantly moderate the effects. CONCLUSION: Cognitive-aerobic exercise had the most significant effect on improving working memory in children with ADHD. Higher intervention frequency and longer intervention duration may enhance its effects. Future research should explore the impact of these factors and consider increasing sample sizes to validate the role of these moderators. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=627915.