diff --git a/rag2_0/tool/ModelTool.py b/rag2_0/tool/ModelTool.py
index be9daad..7973268 100755
--- a/rag2_0/tool/ModelTool.py
+++ b/rag2_0/tool/ModelTool.py
@@ -167,11 +167,9 @@ class OpenAiLLM:
 
         self._kwargs = kwargs
 
-    def invoke(self, user_prompt="你是谁？", need_retry=True, api_key:str = None, **extra_kwargs):
+    def invoke(self, user_prompt="你是谁？", api_key:str = None, **extra_kwargs):
         # 初始化 OpenAI 客户端
 
-        max_retries = 3
-        retry_count = 0
         # 合并额外的kwargs与self._kwargs
         kwargs = {**self._kwargs}
         if extra_kwargs:
@@ -183,44 +181,21 @@ class OpenAiLLM:
         if api_key is None:
             api_key = APIKeyManager.get_api_key()  
             
-        if need_retry:
-            while retry_count < max_retries:
-                try:
-                    
-                    # 使用with语句创建客户端，确保资源会被正确释放
-                    with OpenAI(api_key=api_key, base_url=self._url) as client:
-                        # 创建 Completion 请求. 超时120s
-                        completion = client.chat.completions.create(
-                            model=self._model,
-                            messages=[{'role': 'user', 'content': user_prompt}],
-                            **self._kwargs
-                        )
-                        return completion.choices[0].message
-                    
-                except Exception as e:
-                    retry_count += 1
-                    if retry_count == max_retries:
-                        raise RuntimeError(f"OpenAiLLM:invoke:error:{str(e)}") from e
-                    else:
-                        time.sleep(5*retry_count)  # 重试前等待5秒*重试次数
-        else:
-            try:
-                # 创建 Completion 请求. 超时120s
-                # 使用with语句创建客户端，确保资源会被正确释放
-                with OpenAI(api_key=api_key, base_url=self._url) as client:
-                    completion = client.chat.completions.create(
-                        model=self._model,
-                        messages=[{'role': 'user', 'content': user_prompt}],
-                        **self._kwargs
-                    )
-                    return completion.choices[0].message
-            except Exception as e:
-                raise RuntimeError(f"OpenAiLLM:invoke:error:{str(e)}") from e
+        try:
+            # 创建 Completion 请求. 超时120s
+            # 使用with语句创建客户端，确保资源会被正确释放
+            with OpenAI(api_key=api_key, base_url=self._url) as client:
+                completion = client.chat.completions.create(
+                    model=self._model,
+                    messages=[{'role': 'user', 'content': user_prompt}],
+                    **self._kwargs
+                )
+                return completion.choices[0].message
+        except Exception as e:
+            raise RuntimeError(f"OpenAiLLM:invoke:error:{str(e)}") from e
 
     async def ainvoke(self, user_prompt="你是谁？", **extra_kwargs):
         """异步调用OpenAI API"""
-        max_retries = 3
-        retry_count = 0
         
         # 合并额外的kwargs与self._kwargs
         kwargs = {**self._kwargs}