Adequate and timely antibiotic therapy is crucial for the treatment of sepsis. Innovative systems, like the Q-linea ASTar, have been developed to perform rapid antimicrobial susceptibility testing (AST) directly from positive blood cultures (BCs). We conducted a prospective study to evaluate ASTar under real-life conditions with a focus on time-to-result and impact on antimicrobial therapy. Over 2 months, all positive BCs that showed Gram-negative rods upon microscopy were tested with the ASTar and our standard procedure (VITEK 2 from short-term culture). Additionally, we included multidrug-resistant Gram-negative bacteria from our archive. Both methods were compared to broth microdilution. In total, 78 bacterial strains (51 prospective and 27 archived) were tested. ASTar covered 94% of the species encountered. The categorical and essential agreement was 95.6% and 90.7%, respectively. ASTar caused 2.4% minor, 2.0% major, and 2.4% very major errors. The categorical agreement was similar to standard procedure. The average time between BC sampling and the availability of the antibiogram for the attending physician was 28 h 49 min for ASTar and 44 h 18 min for standard procedure. ASTar correctly identified all patients who required an escalation of antimicrobial therapy and 75% of those who were eligible for de-escalation. In conclusion, ASTar provided reliable AST results and significantly shortened the time to obtain an antibiogram. However, the percentage of patients that will profit from ASTar in a low-resistance setting is limited, and it is currently unclear if a change of therapy 29 h after BC sampling will have a significant impact on the patient's prognosis.