Objective: The aim of this systematic review was to evaluate the effects of tai chi on metabolic control and body composition indicators in patients with type 2 diabetes mellitus. Design: Systematic review and meta-analysis of existing literature. Methods: Electronic resource databases were searched to collect eligible studies. Two reviewers selected studies and independently evaluated methodological quality. Results: Twenty-three studies were included in this meta-analysis. The pooled results showed that tai chi had significant effects in improving metabolic indices, such as fasting blood glucose (mean difference (MD) = –1.04; 95% confidence interval (95% CI) –1.42 to 0.66; p < 0.01) and total cholesterol (MD = –0.50; 95% CI –0.86 to –0.13; p < 0.01) compared with conventional clinical therapy. Most indices did not support the use of tai chi over aerobic exercise, except for glycated haemoglobin (HbA1c) (MD = –0.24; 95% CI –0.49 to 0.00; p < 0.01) and high-density lipoprotein (MD = 0.07; 95% CI 0.01 to 0.12; p < 0.01). Conclusion: Tai chi had better effects on metabolic control and body composition indicators than clinical conventional therapy, but only on HbA1c and HDL were superior than that of aerobic exercise. The best time-window for tai chi intervention may differ with different metabolic indices.