AimTo evaluate the efficacy and harms of a short (7–10 days) compared with a standard (10–14 days) duration of antibiotics in culture‐proven neonatal sepsis for reducing all‐cause mortality, treatment failure and duration of hospitalisation.MethodsMedline, EMBASE and Cochrane CENTRAL were searched for randomised trials.ResultsWe included five studies, all conducted in India (447 infants with a gestational age greater than 32 weeks). Except for one study, all studies were at high risk of bias. All‐cause mortality was reported in three studies with only one death reported in the standard duration regimen arm (243 patients, very low certainty). A meta‐analysis showed no evidence of the effect on treatment failure (RR of 1.47 [95% CI 0.48–4.50], 440 patients, five studies, very low certainty) of short‐term antibiotics. Short‐term antibiotic regimen shortened the duration of hospitalisation by 4 days (mean difference of −4.04 days [95% CI −5.47 to −2.61]; 4 studies; 371 patients; very low certainty).ConclusionAmong studies focused on infants born with a gestational age greater than 32 weeks, short‐term administration of antibiotics may shorten the duration of hospitalisation, but the evidence is very uncertain. The evidence on other predefined outcomes is very uncertain to draw definite conclusions.