This study aimed to evaluate the validation of the leptin receptor-deficient mice model for secondary osteoporosis associated with type 2 diabetes mellitus (T2DM) at bone micro-architectural level. Thirty three 36-week old male mice were divided into four groups: normal control (db/m) (n = 7), leptin receptor-deficient T2DM (db/db) (n = 8), human C-reactive protein (CRP) transgenic normal control (crp/db/m) (n = 7), and human CRP transgenic T2DM (crp/db/db) (n = 11). Lumber vertebrae (L5) and bilateral lower limbs were scanned by micro-CT to analyze trabecular and cortical bone quality. Right femora were used for three-point bending to analyze the mechanical properties. Trabecular bone quality at L5 was better in db/db or crp/db/db group in terms of bone mineral density (BMD), bone volume fraction, connectivity density, trabecular number and separation (all p < 0.05). However the indices measured at proximal tibia showed comparable trabecular BMD and microarchitecture among the four groups. Femur length in crp/db/db group was significantly shorter than db/m group (p < 0.05) and cortices were thinner in db/db and crp/db/db groups (p > 0.05). Maximum loading and energy yield in mechanical test were similar among groups while the elastic modulus in db/db and crp/db/db significantly lower than db/m. The leptin-receptor mice is not a proper model for secondary osteoporosis associated with T2DM.