According to A Randomized Trial of Unruptured Brain Arteriovenous Malformations (ARUBA), conservative treatment seems to be superior to any intervention for unruptured brain arteriovenous malformations (AVMs). This study aims to evaluate safety and efficacy of upfront and repeated Gamma-Knife Radiosurgery (GKRS) in patients harboring small AVMs fulfilling the inclusion criteria of ARUBA. A retrospective study was conducted to evaluate outcomes of unruptured naive brain AVM with a volume ⩽ 5 mL eligible to ARUBA treated by GKRS with at least 3 years of follow-up. From 1992 to 2014, 249 patients fulfilled the inclusion criteria of this study. The median age was 36 years (range: 18-78 years). The median treated volume of the nidus was 1.3 mL (range: 0.4-5 mL) and 63% of the AVM were in eloquent areas. Radiosurgery-based AVM score was 1-1.8 (76%), the Spetzler-Martin grade was II-III (73%), and the Virginia Radiosurgery AVM scale was ≤1 point (75%). The overall AVM obliteration rate was 77.1% after up to 3 GKRS sessions. The median dose at the margin was 24 Gy (15-25 Gy) and the median follow-up was 45 months (range: 36-205 months). Eight patients (3.2%) experienced hemorrhage after GKRS, corresponding to a post-GKRS hemorrhage annual rate of 1.03%. Permanent symptomatic radio-induced changes rate was 2% (4 increased seizures, 1 neurologic deficit). The very low toxicity rate and the high occlusion rate suggest in favor of upfront and repeated GKRS for unruptured small AVMs (⩽5 mL).