Diabetes mellitus is a complex metabolic disorder characterized by chronic hyperglycemia, hypoinsulinemia, and ketosis. To access the biochemical process of diabetes, we applied quantitative ¹H NMR-based metabonomics to analyze urine, serum, and liver extracts from streptozotocin-induced diabetic rats. Principle component analysis (PCA) of ¹H NMR spectra disclosed metabolic pattern differences between diabetic and control rats, and identified the related metabolic changes. The PCA scores plot demonstrated that the diabetic group could be distinguished from the control group, indicating that the metabolic characteristics of the two groups were markedly different. Our work reveals the accumulation of triglycerides, fatty acids and acetoacetate in diabetic rats, and may provide an efficient, convenient way for evaluating the pathological state and biochemical process of diabetes mellitus.