Biomarkers indicative of intracerebral hemorrhage (ICH) may help triage acute stroke patients in the pre-hospital phase. We hypothesized that serum concentration of glial fibrillary acidic protein (GFAP) in combination with ubiquitin carboxy-terminal hydrolase-L1 (UCH-L1), measured by a rapid bio-assay, could be used to distinguish ICH from ischemic stroke.This prospective two-center study recruited patients with a clinical diagnosis of acute stroke both in the pre-hospital phase and at hospital admission (within 4 and 6 h after symptom onset, respectively). Blood samples were analyzed for concentrations of GFAP and UCH-L1 using ELISA techniques. The reference standard was the diagnosis of ICH, ischemic stroke, or stroke mimicking condition achieved after clinical workup including brain imaging.A total of 251 patients were included (mean age [± SD] 72 ± 15 years; 5 ICH, 23 ischemic strokes and 14 stroke mimics in the pre-hospital part; and 59 ICH, 148 ischemic strokes and 2 stroke mimics in the in-hospital part). Mean delay (± SD) from symptom onset to blood withdrawal was 130 ± 79 min for the pre-hospital patients and 136 ± 86 min for the in-hospital patients. Both GFAP and UCH-L1 serum concentrations were higher in patients having ICH as compared to other diagnoses (GFAP: median 330 ng/L [interquartile range 64-7060, range 8-56,100] vs. 27.5 ng/L [14-57.25, 0-781], p < 0.001; UCH-L1: 401 ng/L [265-764, 133-1812] vs. 338 ng/L [213-549.5, 0-2950], p = 0.025). Area-under-the-curve values were 0.866 (95% CI 0.809-0.924, p < 0.001) for GFAP, and 0.590 (0.511-0.670, p = 0.033) for UCH-L1. Regarding overall diagnostic accuracy, UCH-L1 did not add significantly to the performance of GFAP.GFAP may differentiate ICH from ischemic stroke and stroke mimics. A point-of-care test to distinguish between ischemic and hemorrhagic strokes might facilitate triage to different treatment pathways or locations, or be used to select patients for trials of ultra-early interventions.