Diabetic Peripheral Neuropathy (DPN) typically is accompanied by painful symptoms. Several therapeutic agents have been tried for symptomatic relief, but with varying results. The use of non-invasive neuromodulation (NINM) is a potential treatment option for DPN. The objective of our study is to evaluate NINM effects on pain rating and nerve conduction velocity in DPN patients. The search was carried out in seven databases until Aug 30th, 2019. Finally, twenty studies met the inclusion criteria. We found a significant reduction of pain scores by central NINMs (effect size [ES] = − 0.75, 95% CI = − 1.35 to − 0.14), but not by the overall peripheral techniques (electrical and electromagnetic) (ES = − 0.58, 95% CI = − 1.23 to 0.07). However, the subgroup of peripheral electrical NINMs reported a significant higher effect (ES = − 0.84, 95% CI = − 1.57 to − 0.11) compared to electromagnetic techniques (ES = 0.21; 95% CI = − 1.00 to 1.42, I2 = 95.3%). Other subgroup analysis results show that NINMs effects are higher with intensive protocols and in populations with resistant symptoms or intolerance to analgesic medications. Besides, NINMs can increase motor nerves velocity (ES = 1.82; 95% CI = 1.47 to 2.17), and there were no effects on sensory nerves velocity (ES = 0.01, 95% CI = − 0.79 to 0.80). The results suggest that central and peripheral electrical NINMs could reduce neuropathic pain among DPN patients, without reported adverse events. Well-powered studies are needed to confirm that NINM techniques as an alternative effective and safe treatment option.