Objective: To quantitatively evaluate the degree of renal cortical fat infiltration in patients with type 2 diabetes mellitus (T2DM) using magnetic resonance iterative decomposition of water and fat with echo asymmetry and least-squares estimation-iron quantification (IDEAL-IQ) technology, analyze its quantitative correlation with different grades of insulin resistance (IR), and explore the mediating effect of renal tubular injury between them. Methods: A total of 124 patients diagnosed with T2DM from July 2022 to October 2025 were prospectively enrolled. According to the homeostasis model assessment of insulin resistance (HOMA-IR), the patients were divided into the mild IR group (HOMA-IR 2.5-5.0, n=42), moderate IR group (HOMA-IR 5.1~10.0, n=40) and severe IR group (HOMA-IR>10.0, n=42). All subjects underwent renal 3.0T MRI-IDEAL-IQ examination to quantitatively measure the renal cortical fat fraction (FF). Glycometabolic indicators including fasting plasma glucose (FPG) and glycosylated hemoglobin (HbA1c), renal function indicators including serum creatinine (Scr) and estimated glomerular filtration rate (eGFR), and renal tubular injury markers including urine neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1) and urine albumin/creatinine ratio (UACR) were detected, and HOMA-IR was calculated. The differences in renal cortical FF, renal tubular injury markers and renal function indicators among different IR grade groups were compared. Pearson correlation analysis was used to explore the correlation between renal cortical FF and various indicators. Multiple linear regression analysis was performed to verify the independent effect of IR on renal cortical FF. Mediation effect analysis was adopted to investigate the mediating role of renal tubular injury between IR and renal cortical fat infiltration. Results: With the aggravation of IR grading, renal cortical FF, NGAL, KIM-1, UACR and Scr in patients showed a gradual increasing trend, while eGFR showed a gradual decreasing trend, with statistically significant differences among all grade groups (P < 0.05). Pearson correlation analysis revealed that renal cortical FF was highly positively correlated with HOMA-IR, KIM-1, UACR, and the comprehensive score of renal tubular injury (r=0.834, P<0.001; r=0.810, P < 0.001; r=0.834, P < 0.001; r=0.831, P < 0.001), moderately positively correlated with NGAL, Scr, and BUN (r=0.673-0.772, all P < 0.001), and highly negatively correlated with eGFR (r=-0.817, P < 0.001). Multiple linear regression analysis indicated that after adjusting for age, gender, body mass index (BMI), and HbA1c, HOMA-IR was still an independent risk factor for the increase of renal cortical FF in T2DM patients (β=0.419, P < 0.001). Mediation effect analysis showed that renal tubular injury (comprehensive score of NGAL+KIM-1+UACR) had a partial mediation effect between IR and cortical fat infiltration, accounting for 45.98% of the total effect. Conclusion: The degree of renal cortical fat infiltration in T2DM patients has a significant dose-effect relationship with IR grade. IR directly promotes renal cortical fat deposition, and also indirectly mediates the occurrence and development of renal cortical fat infiltration by inducing renal tubular injury. Renal cortical FF quantitatively detected by IDEAL-IQ technology can be used as a non-invasive imaging indicator to reflect the degree of IR and early renal tubular injury in T2DM patients, providing a new perspective for the early assessment of diabetic renal complications.