The application of high cutoff (HCO) membranes for continuous renal replacement therapy remains unclear in septic acute kidney injury (S-AKI) patients.S-AKI patients who received continuous veno-venous hemodiafiltration (CVVHDF) were randomly assigned to the experimental group (HCO membrane) and the control group (high flux membrane, HF membrane). Interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) in serum and waste fluid were measured at 0, 2, 12, and 24 h after CVVHDF initiation and the 28-day mortality.Eleven patients were randomized to the HCO group, and 9 patients in the HF group, with a mean age of 54.9 ± 3.2 years and 6 patients (30%) being female. After 24 h of treatment with CVVHDF, there were significant reductions in serum IL-6 and TNF-α concentrations in the HCO group (p = 0.001, 0.015) and HF group (p = 0.004, 0.031). The serum IL-6 reduction rate of the HCO group was significantly higher than that of the HF group (79.21% vs. 42.69%, p = 0.025), while serum TNF-α reduction rates were comparable between the 2 groups. There were no significant changes in serum albumin after 24 h using either HCO membrane (28.7 ± 1.7 g/L vs. 32.7 ± 1.6 g/L, p = 0.138) or HF membrane (29.6 ± 1.1 g/L vs. 32.6 ± 1.3 g/L, p = 0.055). The two groups had similar 24-h filter clotting rates and 28-day mortality.While CVVHDF with the HCO membrane and HF membrane both achieved significant reductions in serum cytokine levels, the HCO membrane was associated with a greater reduction rate in IL-6 but not in TNF-α. No difference was observed in serum albumin, mortality, or filter clotting.Registry number: ChiCTR2000039725.© 2025 International Center for Artificial Organ and Transplantation (ICAOT) and Wiley Periodicals LLC.