diff --git a/python/llm/src/ipex_llm/transformers/convert.py b/python/llm/src/ipex_llm/transformers/convert.py
index 76d13ccb..b7899d73 100644
--- a/python/llm/src/ipex_llm/transformers/convert.py
+++ b/python/llm/src/ipex_llm/transformers/convert.py
@@ -1691,11 +1691,13 @@ def _optimize_post(model, lightweight_bmm=False):
         module = importlib.import_module(modeling_module_name)
         from ipex_llm.transformers.models.common import rms_norm_forward
         from ipex_llm.transformers.models.qwen2 import qwen2_mlp_forward
+        from ipex_llm.transformers.models.qwen2_vl import qwen2_vision_get_dtype
         from ipex_llm.transformers.models.qwen2_vl import qwen2_vision_attention_forward
         from ipex_llm.transformers.models.qwen2_vl import qwen2_vl_model_forward
         from ipex_llm.transformers.models.qwen2_vl import qwen2_vl_attention_forward
         convert_forward(model, module.Qwen2RMSNorm, rms_norm_forward)
         convert_forward(model, module.Qwen2MLP, qwen2_mlp_forward)
+        model.visual.get_dtype = MethodType(qwen2_vision_get_dtype, model.visual)
         convert_forward(model, module.VisionAttention, qwen2_vision_attention_forward)
         convert_forward(model, module.Qwen2VLModel, qwen2_vl_model_forward)
         convert_forward(model, module.Qwen2VLAttention, qwen2_vl_attention_forward)
diff --git a/python/llm/src/ipex_llm/transformers/models/qwen2_vl.py b/python/llm/src/ipex_llm/transformers/models/qwen2_vl.py
index 81a07ced..ac800291 100644
--- a/python/llm/src/ipex_llm/transformers/models/qwen2_vl.py
+++ b/python/llm/src/ipex_llm/transformers/models/qwen2_vl.py
@@ -175,6 +175,10 @@ def qwen2_vl_model_forward(
     )
 
 
+def qwen2_vision_get_dtype(self) -> torch.dtype:
+    return self.patch_embed.proj.weight.dtype
+
+
 def qwen2_vision_attention_forward(
     self,
     hidden_states: torch.Tensor,