diff --git a/optimum/intel/openvino/modeling_visual_language.py b/optimum/intel/openvino/modeling_visual_language.py
index e895a76e7..9bb3b1f67 100644
--- a/optimum/intel/openvino/modeling_visual_language.py
+++ b/optimum/intel/openvino/modeling_visual_language.py
@@ -1020,6 +1020,10 @@ def preprocess_inputs(
                 prompt = "<image>\n" + text
             else:
                 prompt = text
+
+        if hasattr(processor, "patch_size") and processor.patch_size is None:
+            processor.patch_size = config.vision_config.patch_size
+
         inputs = processor(images=image, text=prompt, return_tensors="pt")
         return inputs