IntroductionTo evaluate the relative efficacy of non-invasive brain stimulation (NIBS) and digital interventions for improving social cognition in autism spectrum disorder (ASD).MethodsA systematic search of PubMed, Embase, Web of Science, Cochrane, CNKI to February 2026 yielded 50 randomized controlled trials (RCTs). A Bayesian network meta-analysis was performed using R, ranking interventions via Surface Under the Cumulative Ranking curve (SUCRA) and assessing evidence certainty with Confidence in Network Meta-Analysis (CINEMA).ResultsAcross the included trials (mean age range: 2.54 to 20.80 years), among 9 interventions, transcranial direct current stimulation (tDCS) emerged as optimal (SUCRA = 88.24%), significantly outperforming treatment as usual (TAU) (SMD=-1.69, 95% CrI: -2.67, -0.70). This was followed by high-frequency repetitive transcranial magnetic stimulation (HF-rTMS) (84.38%, SMD=-1.60, 95% CrI: -2.32, -0.89), low-frequency repetitive transcranial magnetic stimulation (LF-rTMS) (74.22%, SMD=-1.36, 95% CrI: -1.77, -0.96), and theta-burst stimulation (TBS) (64.21%, SMD=-1.19, 95% CrI: -2.02, -0.35); all significantly outperformed TAU. Standalone digital interventions ranked lower. Meta-regression showed no significant age moderating effect.ConclusionstDCS is the most efficacious standalone intervention for ASD social cognitive deficits. We recommend tDCS or TBS clinically and advocate a synergistic paradigm integrating NIBS neural priming with digital training. However, future large-scale, rigorous RCTs are required to further validate both these standalone modalities and the proposed combined protocols.Systematic review registrationhttps://www.crd.york.ac.uk/PROSPERO/view/CRD420261337491, identifier CRD420261337491.