diff --git a/1/model.py b/1/model.py
index 6d615fd..e1b4860 100644
--- a/1/model.py
+++ b/1/model.py
@@ -1,5 +1,4 @@
 import triton_python_backend_utils as pb_utils
-from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig
 import numpy as np
 import json
 
@@ -30,12 +29,13 @@ class TritonPythonModel:
             input_text = self._get_input_value(request, "INPUT")
             self.logger.log_info(f"INPUT 출력:\n{input_text}")
             
-            
-            output = "HELLO_WORLD"
+            random_string = ''.join(random.choice(string.ascii_letters + string.digits) for _ in range(16))
+
+            output = random_string
             self.logger.log_info(f"OUTPUT 출력:\n{output}")
 
             # 생성된 텍스트를 Triton 출력 텐서로 변환합니다.
-            output_tensor = pb_utils.Tensor("text_output", np.array(output.encode('utf-8'), dtype=np.bytes_))
+            output_tensor = pb_utils.Tensor("OUTPUT", np.array(output.encode('utf-8'), dtype=np.bytes_))
 
             # 응답 객체를 생성하고 출력 텐서를 추가합니다.
             responses.append(pb_utils.InferenceResponse(output_tensors=[output_tensor]))