Attention-deficit/hyperactivity disorder (ADHD) is one of the most common neurodevelopmental disorders among children. Pharmacotherapy has been the primary treatment for ADHD, supplemented by behavioral interventions. Digital and exercise interventions are promising nonpharmacologic approaches for enhancing the physical and psychological health of children with ADHD. However, the combined impact of digital and exercise therapies remains unclear. The aim of this study was to determine whether BrainFit, a novel digital intervention combining gamified cognitive and exercise training, is efficacious in reducing ADHD symptoms and executive function (EF) among school-aged children with ADHD. This 4-week prospective randomized controlled trial included 90 children (6-12 years old) who visited the ADHD outpatient clinic and met the diagnostic criteria for ADHD. The participants were randomized (1:1) to the BrainFit intervention (n=44) or a waitlist control (n=46) between March and August 2022. The intervention consisted of 12 30-minute sessions delivered on an iPad over 4 weeks with 3 sessions per week (Monday, Wednesday, and Friday after school) under the supervision of trained staff. The primary outcomes were parent-rated symptoms of attention and hyperactivity assessed according to the Swanson, Nolan, and Pelham questionnaire (SNAP-IV) rating scale and EF skills assessed by the Behavior Rating Inventory of Executive Function (BRIEF) scale, evaluated pre and post intervention. Intention-to-treat analysis was performed on 80 children after attrition. A nonparametric resampling-based permutation test was used for hypothesis testing of intervention effects. Among the 145 children who met the inclusion criteria, 90 consented and were randomized; ultimately, 80 (88.9%) children completed the study and were included in the analysis. The participants' average age was 8.4 (SD 1.3) years, including 63 (78.8%) male participants. The most common ADHD subtype was hyperactive/impulsive (54/80, 68%) and 23 (29%) children had severe symptoms. At the endpoint of the study, the BrainFit intervention group had a significantly larger improvement in total ADHD symptoms (SNAP-IV total score) as compared to those in the control group (β=-12.203, 95% CI -17.882 to -6.523; P<.001), owing to lower scores on the subscales Inattention (β=-3.966, 95% CI -6.285 to -1.647; P<.001), Hyperactivity/Impulsivity (β=-5.735, 95% CI -8.334 to -3.137; P<.001), and Oppositional Defiant Disorder (β=-2.995, 95% CI -4.857 to -1.132; P=.002). The intervention was associated with significant reduction in the Metacognition Index (β=-6.312, 95% CI -10.973 to -1.650; P=.006) and Global Executive Composite (β=-5.952, 95% CI -10.214 to -1.690; P=.003) on the BRIEF. No severe intervention-related adverse events were reported. This novel digital cognitive-physical intervention was efficacious in school-age children with ADHD. A larger multicenter effectiveness trial with longer follow-up is warranted to confirm these findings and to assess the durability of treatment effects. Chinese Clinical Trial Register ChiCTR2300070521; https://www.chictr.org.cn/showproj.html?proj=177806.