Factors that contribute to bone fragility in type 2 diabetes are not well understood. We assessed the effects of intensive glycemic control, thiazolidinediones (TZDs), and A1C levels on bone geometry and strength at the radius and tibia. In a substudy of the Action to Control Cardiovascular Risk in Diabetes trial, peripheral quantitative computed tomographic (pQCT) scans of the radius and tibia were obtained 2 years after randomization on 73 participants (intensive n = 35, standard n = 38). TZD use and A1C levels were measured every 4 months during the trial. Effects of intervention assignment, TZD use, and A1C on pQCT parameters were assessed in linear regression models. Intensive, compared with standard, glycemic control was associated with 1.3 % lower cortical volumetric BMD at the tibia in men (p = 0.02) but not with other pQCT parameters. In women, but not men, each additional year of TZD use was associated with an 11 % lower polar strength strain index (SSIp) at the radius (p = 0.04) and tibia (p = 0.002) in models adjusted for A1C levels. In women, each additional 1 % increase in A1C was associated with an 18 % lower SSIp at the ultradistal radius (p = 0.04) in models adjusted for TZD use. There was no consistent evidence of an effect of intensive, compared with standard, glycemic control on bone strength at the radius or tibia. In women, TZD use may reduce bone strength at these sites. Higher A1C may also be associated with lower bone strength at the radius, but not tibia, in women.