Hyperinsulinemic-euglycemic clamps are considered the "gold standard" for assessing whole body insulin sensitivity. When used in combination with tracer dilution techniques and physiological insulin concentrations, insulin sensitization can be dissected and attributed to hepatic and peripheral (primarily muscle) effects. Non-human primates (NHPs), such as rhesus monkeys, are the closest pre-clinical species to humans, and thus serve as an ideal model for testing of compound efficacy to support translation to human efficacy. We determined insulin infusion rates that resulted in high physiological insulin concentrations that elicited maximal pharmacodynamic responses during hyperinsulinemic-euglycemic clamps. These rates were then used with [U-13C]-D-glucose, to assess and document the degrees of hepatic and peripheral insulin resistance between healthy and insulin-resistant, dysmetabolic NHPs. Next, dysmetabolic NHPs were treated for 28 days with pioglitazone (3 mg/kg) and again had their insulin sensitivity assessed, illustrating a significant improvement in hepatic and peripheral insulin sensitivity. This coincided with a significant increase in insulin clearance, and normalization of circulating adiponectin. In conclusion, we have determined a physiological clamp paradigm (similar to humans) for assessing glucose turnover in NHPs. We have also demonstrated that insulin-resistant, dysmetabolic NHPs respond to the established insulin sensitizer, pioglitazone, thus confirming their use as an ideal pre-clinical translational model to assess insulin sensitizing compounds.