Background Preprocedural predictors of outcome in patients with acute basilar artery occlusion (ABAO) who have undergone endovascular treatment (EVT) remain controversial. Our aim was to determine if pre-EVT diffusion-weighted imaging cerebellar infarct volume (CIV) is a predictor of 90-day outcomes. Methods We analyzed consecutive MRI-selected endovascularly treated patients with ABAO within the first 24 hours after symptom onset. Successful reperfusion was defined as a modified Thrombolysis in Cerebral Infarction score of 2b–3. Using the initial MRI, baseline CIV was calculated in mL on an apparent diffusion coefficient map reconstruction (Olea Sphere software). CIV was analyzed in univariate and multivariable models as a predictor of 90-day functional independence (modified Rankin Scale (mRS) 0–2) and mortality. According to receiver operating characteristic (ROC) analysis, the optimal cut-off was determined by maximizing the Youden index to evaluate the prognostic value of CIV. Results Of the 110 MRI-selected patients with ABAO, 64 (58.18%) had a cerebellar infarct. The median CIV was 9.6 mL (IQR 2.7–31.4). Successful reperfusion was achieved in 81.8% of the cases. At 90 days the proportion of patients with mRS ≤2 was 31.8% and the overall mortality rate was 40.9%. Baseline CIV was significantly associated with 90-day mRS 0–2 (p=0.008) in the univariate analysis and was an independent predictor of 90-day mortality (adjusted OR 1.79, 95% CI 1.25 to 2.54, p=0.001). The ROC analysis showed that a CIV ≥4.7 mL at the initial MRI was the optimal cut-off to discriminate patients with a higher risk of death at 90 days (area under the ROC curve (AUC)=0.74, 95% CI 0.61 to 0.87, sensitivity and specificity of 87.9% and 58.1%, respectively). Conclusions In our series of MRI-selected patients with ABAO, pre-EVT CIV was an independent predictor of 90-day mortality. The risk of death was increased for baseline CIV ≥4.7 mL.