撤回之前使用langchain_openai调用模型的逻辑。因为暂时无法解决调用Qwen3禁用思考模式的问题

rag2_0/tool/ModelTool.py

@@ -217,7 +217,7 @@ class OpenAiLLM:
             except Exception as e:
                 raise RuntimeError(f"OpenAiLLM:invoke:error:{str(e)}.api_key:{api_key}") from e
 
-    async def invoke_async(self, user_prompt="你是谁？", need_retry=True, **extra_kwargs):
+    async def ainvoke(self, user_prompt="你是谁？", **extra_kwargs):
         """异步调用OpenAI API"""
         max_retries = 3
         retry_count = 0
@@ -231,38 +231,17 @@ class OpenAiLLM:
             timeout = httpx.Timeout(300.0) 
             kwargs["timeout"] = timeout
 
-        if need_retry:
-            while retry_count < max_retries:
-                try:
-                    api_key = APIKeyManager.get_api_key()  
-                    # 使用异步客户端
-                    async with AsyncOpenAI(api_key=api_key, base_url=self._url) as client:
-                        # 创建异步Completion请求
-                        completion = await client.chat.completions.create(
-                            model=self._model,
-                            messages=[{'role': 'user', 'content': user_prompt}],
-                            **kwargs
-                        )
-                        return completion.choices[0].message
-                    
-                except Exception as e:
-                    retry_count += 1
-                    if retry_count == max_retries:
-                        raise RuntimeError(f"OpenAiLLM:invoke_async:error:{str(e)}.api_key:{api_key}") from e
-                    else:
-                        await asyncio.sleep(5*retry_count)  # 异步等待
-        else:
-            try:
-                api_key = APIKeyManager.get_api_key()  
-                async with AsyncOpenAI(api_key=api_key, base_url=self._url) as client:
-                    completion = await client.chat.completions.create(
-                        model=self._model,
-                        messages=[{'role': 'user', 'content': user_prompt}],
-                        **kwargs
-                    )
-                    return completion.choices[0].message
-            except Exception as e:
-                raise RuntimeError(f"OpenAiLLM:invoke_async:error:{str(e)}.api_key:{api_key}") from e
+        try:
+            api_key = APIKeyManager.get_api_key()  
+            async with AsyncOpenAI(api_key=api_key, base_url=self._url) as client:
+                completion = await client.chat.completions.create(
+                    model=self._model,
+                    messages=[{'role': 'user', 'content': user_prompt}],
+                    **kwargs
+                )
+                return completion.choices[0].message
+        except Exception as e:
+            raise RuntimeError(f"OpenAiLLM:ainvoke:error:{str(e)}") from e
 
 if __name__ == "__main__":
     # 测试重排模型
@@ -291,7 +270,7 @@ if __name__ == "__main__":
         
         # 测试异步LLM调用
         llm = OpenAiLLM()
-        response = await llm.invoke_async("你好，请简单介绍一下自己")
+        response = await llm.ainvoke("你好，请简单介绍一下自己")
         print(f"异步LLM响应: {response.content}")
     
     # 如果需要运行异步测试，取消下面的注释