OBJECTIVES: Rapid identification of Acinetobacter species is critical since members of the A. baumannii (Ab) group differ in antibiotic susceptibility and clinical outcomes. A. baumannii, A. pittii and A. nosocomialis can be identified by MALDI-TOF/MS, while the novel species A. seifertii and A. dijkshoorniae cannot. Low identification rates for A. nosocomialis have also been reported. We evaluated the use of MALDI-TOF/MS to identify isolates of A. seifertii and A. dijkshoorniae and revisited the identification of A. nosocomialis to update the Bruker taxonomy database. METHODS: Species characterisation was performed by rpoB-clustering and MLSA. MALDI-TOF/MS spectra were recovered from formic acid/acetonitrile bacterial extracts overlaid with alpha-cyano-4-hydroxy-cinnamic acid matrix on a MicroflexLT in linear positive mode and 2,000-20,000 m/z range mass. Spectra were examined with the ClinProTools v2.2 software. Mean spectra (MSP) were created with the BioTyper software. RESULTS: Seventy-eight Acinetobacter isolates representative of the Ab group were used to calculate the average spectra/species and generate pattern recognition models. Species-specific peaks were identified for all species, and MSPs derived from 3 A. seifertii, 2 A. dijkshoorniae and 2 A. nosocomialis strains were added to the Bruker taxonomy database, allowing successful identification of all isolates using spectra from either bacterial extracts or direct colonies, resulting in a positive predictive value (PPV) of 99.6% (777/780) and 96.8% (302/312), respectively. CONCLUSIONS: The use of post-processing data software identified statistically significant species-specific peaks to generate reference signatures for rapid accurate identification of species within the Ab group, providing relevant information for the clinical management of Acinetobacter infections.