diff --git a/vllm/model_executor/models/blip2.py b/vllm/model_executor/models/blip2.py
index 2f2b880bb0e14..ed98a3008c567 100644
--- a/vllm/model_executor/models/blip2.py
+++ b/vllm/model_executor/models/blip2.py
@@ -560,8 +560,8 @@ class Blip2ForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsPP,
         self.make_empty_intermediate_tensors = (
             self.language_model.make_empty_intermediate_tensors)
 
-    def _create_image_input(self,
-                            **kwargs: object) -> Optional[Blip2ImageInputs]:
+    def _parse_and_validate_image_input(
+            self, **kwargs: object) -> Optional[Blip2ImageInputs]:
         pixel_values = kwargs.pop("pixel_values", None)
         image_embeds = kwargs.pop("image_embeds", None)