The aim of this study was to investigate the clinical efficacy of a combination of nirmatrelvir and ritonavir (NMV‐r) for treating COVID‐19 in patients with diabetes mellitus (DM). This retrospective cohort study used the TriNetX research network to identify adult diabetic patients with COVID‐19 between January 1, 2020, and December 31, 2022. Propensity score matching was used to match patients who received NMV‐r (NMV‐r group) with those who did not receive NMV‐r (control group). The primary outcome was all‐cause hospitalization or death during the 30‐day follow‐up period. Two cohorts comprising 13 822 patients with balanced baseline characteristics were created using propensity score matching. During the follow‐up period, the NMV‐r group had a lower risk of all‐cause hospitalization or death than the control group (1.4% n = 193 vs. 3.1% n = 434; hazard ratio HR, 0.497; 95% confidence interval CI, 0.420–0.589). Compared with the control group, the NMV‐r group also had a lower risk of all‐cause hospitalization (HR, 0.606; 95% CI, 0.508–0.723) and all‐cause mortality (HR, 0.076; 95% CI, 0.033–0.175). This lower risk was consistently observed in almost all subgroup analyses, which examined sex (male: 0.520 0.401–0.675; female: 0.586 0.465–0.739), age (age 18–64 years: 0.767 0.601–0.980; ≥65 years: 0.394 0.308–0.505), level of HbA1c (<7.5%: 0.490 0.401–0.599; ≥7.5%: 0.655 0.441–0.972), unvaccinated (0.466 0.362–0.599), type 1 DM (0.453 0.286–0.718) and type 2 DM (0.430 0.361–0.511). NMV‐r can help reduce the risk of all‐cause hospitalization or death in nonhospitalized patients with DM and COVID‐19.