Abstract Background Klebsiella pneumoniae producing KPC variants conferring resistance to ceftazidime–avibactam often remain susceptible to meropenem, suggesting a potential therapeutic use of this antibiotic. Objectives In this study, the efficacy of clinically relevant concentrations of meropenem was evaluated against high-risk clones of ceftazidime–avibactam-resistant K. pneumoniae strains producing KPC variants, in a tandem in vitro time-kill/in vivo Galleria mellonella survival model. Methods In vitro/in vivo efficacy of meropenem against ceftazidime–avibactam-resistant K. pneumoniae of CG16, CG25 and CG258, producing KPC-31, KPC-33, KPC-90, KPC-106 and KPC-114 variants, was evaluated using EUCAST dosing recommendation adjusted to the G. mellonella model. For in vivo assays, untreated, meropenem (40 mg/kg × 1)-treated and ceftazidime–avibactam (40 mg/kg ceftazidime–10 mg/kg avibactam × 1)-treated groups were established, with 60 larvae per group. Kaplan–Meier curves, log-rank tests, univariate Cox regression and hazard ratios (HR) were used to assess treatment effects (P < 0.05). Results For all KPC-variant producers, time-kill assays showed >3 log-kills reduction (−6.91 ± 1.28 SD) after 6 h interaction when exposed to 8–32 mg/L meropenem MIC values (i.e. ≥ × 4 MIC). In the assessment of in vivo efficacy of meropenem, at the 4-day follow-up, mortality rates were 96.7% (untreated), 83.3% (ceftazidime–avibactam-treated) and 13.3% (meropenem-treated) (P < 0.05). Univariate Cox regression analysis showed significantly lower risk in the meropenem group compared to untreated group [HR 0.02 (95% CI: 0.01–0.05)]. Conclusions These pre-clinical results might support use of meropenem as a potential alternative for treatment of infections due to KPC-variant producers displaying in vitro susceptibility to meropenem.