Fix model name issue

Signed-off-by: Jael Gu <mengjia.gu@zilliz.com>

auto_transformers.py

@@ -81,12 +81,11 @@ class AutoTransformers(NNOperator):
         if device is None:
             device = 'cuda' if torch.cuda.is_available() else 'cpu'
         self.device = device
-
-        model_list = self.supported_model_names()
-        assert model_name in model_list, f"Invalid model name: {model_name}. Supported model names: {model_list}"
         self.model_name = model_name
 
         if self.model_name:
+            model_list = self.supported_model_names()
+            assert model_name in model_list, f"Invalid model name: {model_name}. Supported model names: {model_list}"
             self.model = Model(
                 model_name=self.model_name, device=self.device, checkpoint_path=checkpoint_path)
             if tokenizer is None: