In order to evaluate the in vivo thrombogenicity of injectable agents, a suitable animal model is needed. We introduce an ultrasound-guided non-selective cerebral artery occlusion model via the common carotid arteries of rabbits. A total of 30 rabbits were assigned to an experimental group (n = 20) and a control group (n = 10). Each group received 2 mL suspension of embolic agent or 2 mL of normal saline, respectively, under ultrasound guidance. The animals were observed for immediate reaction and underwent magnetic resonance imaging (MRI) scan. Follow-up neurologic examination was conducted 24 h following the procedure. In 7 of the 30 rabbits, 2 in the control group and 5 in the experimental group, the administration of either normal saline or the embolic agent failed. Among the successfully injected 15 experimental animals, 14 showed neurologic impairment or deceased, whereas 1 animal did not show significant neurologic deficit. The MRI of 4 experimental animals showed detectable cerebral infarction on diffusion-weighted imaging. None of the 8 control animals showed neurologic abnormality and their brain MRI was normal. Our minimally invasive model is technically feasible and competent to show thrombogenecity of an injectable agent and consequent in vivo neurologic outcome.