Glucose-stimulated insulin secretion is a multistep process dependent on β-cell metabolic flux. Our previous studies on intact pancreatic islets used two-photon NAD(P)H imaging as a quantitative measure of the combined redox signal from NADH and NADPH (referred to as NAD(P)H). These studies showed that pyruvate, a non-secretagogue, enters β-cells and causes a transient rise in NAD(P)H. To further characterize the metabolic fate of pyruvate, we have now developed one-photon flavoprotein microscopy as a simultaneous assay of lipoamide dehydrogenase (LipDH) autofluorescence. This flavoprotein is in direct equilibrium with mitochondrial NADH. Hence, a comparison of LipDH and NAD(P)H autofluorescence provides a method to distinguish the production of NADH, NADPH, or both. Using this method, the glucose dose response is consistent with an increase in both NADH and NADPH. In contrast, the transient rise in NAD(P)H observed with pyruvate stimulation is not accompanied by a significant change in LipDH, which indicates that pyruvate raises cellular NADPH without raising NADH. In comparison, methyl pyruvate stimulated a robust NADH and NADPH response. These data provide new evidence that exogenous pyruvate does not induce a significant rise in mitochondrial NADH. This inability likely results in its failure to produce the ATP necessary for stimulated secretion of insulin. Overall, these data are consistent with either a restricted pyruvate dehydrogenase-dependent metabolism or a buffering of the NADH response by other metabolic mechanisms.