diff --git a/src/llamafactory/chat/hf_engine.py b/src/llamafactory/chat/hf_engine.py
index 87d9c451..909f8161 100644
--- a/src/llamafactory/chat/hf_engine.py
+++ b/src/llamafactory/chat/hf_engine.py
@@ -166,7 +166,7 @@ class HuggingfaceEngine(BaseEngine):
 
         mm_inputs = template.mm_plugin.get_mm_inputs(**mm_input_dict, seqlens=[prompt_length], processor=processor)
         for key, value in mm_inputs.items():
-            if isinstance(value, list) and all(isinstance(v, torch.Tensor for v in value)):  # for pixtral inputs
+            if isinstance(value, list) and all(isinstance(v, torch.Tensor) for v in value):  # for pixtral inputs
                 value = torch.stack(value)  # assume they have same sizes
             elif not isinstance(value, torch.Tensor):
                 value = torch.tensor(value)