Type 2 diabetes mellitus (T2DM) is a chronic disorder that is characterised by insulin resistance and hyperglycaemia, which over time may give rise to vascular complications. Resveratrol is a plant-derived nutritional supplement shown to have anti-diabetic properties in many animal models. Less evidence is available on its safety and efficacy in the management of T2DM in humans. To assess the efficacy and safety of resveratrol formulations for adults with type 2 diabetes mellitus. We searched the Cochrane Central Register of Controlled Trials, MEDLINE, PubMed, Embase, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), and International Pharmaceutical Abstracts, as well as the International Clinical Trials Registry Platform (ICTRP) Search Portal and ClinicalTrials.gov. The date of the last search was December 2018 for all databases. No language restrictions were applied. All randomised controlled trials (RCTs) comparing effects of oral resveratrol (any dose or formulation, duration, or frequency of administration) with placebo, no treatment, other anti-diabetic medications, or diet or exercise, in adults with a diagnosis of T2DM. Two review authors independently identified and included RCTs, assessed risk of bias, and extracted study-level data. Study authors were contacted for any missing information or for clarification of reported data. We assessed studies for certainty of the evidence using the GRADE instrument. We identified three RCTs with a total of 50 participants. Oral resveratrol not combined with other plant polyphenols was administered at 10 mg, 150 mg, or 1000 mg daily for a period ranging from four weeks to five weeks. The comparator intervention was placebo. Overall, all three included studies had low risk of bias. None of the three included studies reported long-term, patient-relevant outcomes such as all-cause mortality, diabetes-related complications, diabetes-related mortality, health-related quality of life, or socioeconomic effects. All three included studies reported that no adverse events were observed, indicating that no deaths occurred (very low-quality evidence for adverse events, all-cause mortality, and diabetes-related mortality). Resveratrol versus placebo showed neutral effects for glycosylated haemoglobin A1c (HbA1c) levels (mean difference (MD) 0.1%, 95% confidence interval (CI) -0.02 to 0.2; P = 0.09; 2 studies; 31 participants; very low-certainty evidence). Due to the short follow-up period, HbA1c results have to be interpreted cautiously. Similarly, resveratrol versus placebo showed neutral effects for fasting blood glucose levels (MD 2 mg/dL, 95% CI -2 to 7; P = 0.29; 2 studies; 31 participants), and resveratrol versus placebo showed neutral effects for insulin resistance (MD -0.35, 95% CI -0.99 to 0.28; P = 0.27; 2 studies; 36 participants). We found eight ongoing RCTs with approximately 800 participants and two studies awaiting assessment, which, when published, could contribute to the findings of this review. Currently, research is insufficient for review authors to evaluate the safety and efficacy of resveratrol supplementation for treatment of adults with T2DM. The limited available research does not provide sufficient evidence to support any effect, beneficial or adverse, of four to five weeks of 10 mg to 1000 mg of resveratrol in adults with T2DM. Adequately powered RCTs reporting patient-relevant outcomes with long-term follow-up periods are needed to further evaluate the efficacy and safety of resveratrol supplementation in the treatment of T2DM.