ABSTRACT Objective To study the efficacy of Conbercept for the treatment of corneal neovascularization (NV) in a rabbit model. Methods NV was induced by placing sutures. Eight rabbits were used as a control. The other 136 rabbits were randomly divided into two equal groups, and 68 rabbits in each group were divided into four subgroups and given different treatments. Time-course photographs, histological examination, and enzyme-linked immunoassay ELISA analysis for vascular endothelial growth factor were performed at weeks 1, 2, and 3 after injection placement. Results At weeks 1, 2, and 3 after injection placement, there was less expression of corneal NV and VEGF in the conbercept-treated groups than in the saline-treated control groups and less corneal NV and VEGF were expressed in the early treatment group than in the late treatment group. At weeks 2 and 3 after injection, there were fewer corneal NV (length and area) in the early intrastromal injection group with conbercept than in the early subconjunctival injection group with conbercept and a smaller diameter of corneal NV than in the late intrastromal injection group treated with conbercept. Histological examination showed a smaller diameter of corneal NV in all eyes in conbercept-treated groups 1 w after injection than before injection. Treatment with subconjunctival injection with conbercept led to a larger diameter at weeks 2 and 3 than at week 1. Conclusions Subconjunctival and intrastromal administrations of conbercept effectively inhibit corneal NV in rabbits, and the latter has the better effect. The effect is the best in the group with cornea intrastromal injection of conbercept 1 w after suture. Early administration of conbercept may successfully inhibit corneal NV in an animal model.