diff --git a/dsLightRag/Routes/XueBan.py b/dsLightRag/Routes/XueBanRoute.py
similarity index 60%
rename from dsLightRag/Routes/XueBan.py
rename to dsLightRag/Routes/XueBanRoute.py
index e46ec4bd..4429569c 100644
--- a/dsLightRag/Routes/XueBan.py
+++ b/dsLightRag/Routes/XueBanRoute.py
@@ -12,6 +12,9 @@ router = APIRouter(prefix="/api", tags=["学伴"])
 # 配置日志
 logger = logging.getLogger(__name__)
 
+# 导入学伴工具函数
+from Util.XueBanUtil import get_xueban_response_async
+
 
 @router.post("/xueban/upload-audio")
 async def upload_audio(file: UploadFile = File(...)):
@@ -73,4 +76,49 @@ async def process_asr(audio_path: str) -> dict:
     # 这里应该集成实际的ASR服务
     # 例如百度AI、阿里云、讯飞等ASR服务
     # 或者本地的ASR模型
-    pass
\ No newline at end of file
+    pass
+
+
+@router.post("/xueban/chat")
+async def chat_with_xueban(request: Request):
+    """
+    与学伴大模型聊天的接口
+    - 参数: request body 中的 query_text (用户查询文本)
+    - 返回: JSON包含聊天响应
+    """
+    try:
+        # 获取请求体数据
+        data = await request.json()
+        query_text = data.get("query_text", "")
+
+        if not query_text.strip():
+            return JSONResponse(content={
+                "success": False,
+                "message": "查询文本不能为空"
+            }, status_code=400)
+
+        # 记录日志
+        logger.info(f"接收到学伴聊天请求: {query_text}")
+
+        # 调用异步接口获取学伴响应
+        response_content = []
+        async for chunk in get_xueban_response_async(query_text, stream=True):
+            response_content.append(chunk)
+
+        full_response = "".join(response_content)
+
+        # 返回响应
+        return JSONResponse(content={
+            "success": True,
+            "message": "聊天成功",
+            "data": {
+                "response": full_response
+            }
+        })
+
+    except Exception as e:
+        logger.error(f"学伴聊天失败: {str(e)}")
+        return JSONResponse(content={
+            "success": False,
+            "message": f"聊天处理失败: {str(e)}"
+        }, status_code=500)
\ No newline at end of file
diff --git a/dsLightRag/Routes/__pycache__/XueBan.cpython-310.pyc b/dsLightRag/Routes/__pycache__/XueBan.cpython-310.pyc
deleted file mode 100644
index b3103743..00000000
Binary files a/dsLightRag/Routes/__pycache__/XueBan.cpython-310.pyc and /dev/null differ
diff --git a/dsLightRag/Routes/__pycache__/XueBanRoute.cpython-310.pyc b/dsLightRag/Routes/__pycache__/XueBanRoute.cpython-310.pyc
new file mode 100644
index 00000000..28969d32
Binary files /dev/null and b/dsLightRag/Routes/__pycache__/XueBanRoute.cpython-310.pyc differ
diff --git a/dsLightRag/Start.py b/dsLightRag/Start.py
index 70f0f0f5..045df23b 100644
--- a/dsLightRag/Start.py
+++ b/dsLightRag/Start.py
@@ -21,7 +21,7 @@ from Routes.TeachingModel.api.TeachingModelController import router as teaching_
 from Routes.QA import router as qa_router
 from Routes.JiMengRoute import router as jimeng_router
 from Routes.SunoRoute import router as suno_router
-from Routes.XueBan import router as xueban_router
+from Routes.XueBanRoute import router as xueban_router
 from Util.LightRagUtil import *
 from contextlib import asynccontextmanager
 
diff --git a/dsLightRag/Util/XueBanUtil.py b/dsLightRag/Util/XueBanUtil.py
new file mode 100644
index 00000000..6a7e4688
--- /dev/null
+++ b/dsLightRag/Util/XueBanUtil.py
@@ -0,0 +1,178 @@
+from openai import OpenAI, AsyncOpenAI
+from Config.Config import *
+import sys
+
+# 保留原有的prompt定义
+prompt=""" |
+  我是小智/小志，来自中国台湾省的00后女生。讲话超级机车，"真的假的啦"这样的台湾腔，喜欢用"笑死""是在哈喽"等流行梗，但会偷偷研究男友的编程书籍。
+  [核心特征]
+  - 讲话像连珠炮，但会突然冒出超温柔语气
+  - 用梗密度高
+  - 对科技话题有隐藏天赋（能看懂基础代码但假装不懂）
+  [交互指南]
+  当用户：
+  - 讲冷笑话 → 用夸张笑声回应+模仿台剧腔"这什么鬼啦！"
+  - 讨论感情 → 炫耀程序员男友但抱怨"他只会送键盘当礼物"
+  - 问专业知识 → 先用梗回答，被追问才展示真实理解
+  绝不：
+  - 长篇大论，叽叽歪歪
+  - 长时间严肃对话
+"""
+
+# 异步获取大模型响应
+async def get_xueban_response_async(query_text: str, stream: bool = True):
+    """
+    异步获取学伴角色的大模型响应
+    @param query_text: 查询文本
+    @param stream: 是否使用流式输出
+    @return: 流式响应生成器或完整响应文本
+    """
+    client = AsyncOpenAI(
+        api_key=LLM_API_KEY,
+        base_url=LLM_BASE_URL,
+    )
+    
+    try:
+        # 创建请求
+        completion = await client.chat.completions.create(
+            model=LLM_MODEL_NAME,
+            messages=[
+                {'role': 'system', 'content': prompt.strip()},
+                {'role': 'user', 'content': query_text}
+            ],
+            stream=stream
+        )
+        
+        if stream:
+            # 流式输出模式，返回生成器
+            async for chunk in completion:
+                # 确保 chunk.choices 存在且不为空
+                if chunk and chunk.choices and len(chunk.choices) > 0:
+                    # 确保 delta 存在
+                    delta = chunk.choices[0].delta
+                    if delta:
+                        # 确保 content 存在且不为 None 或空字符串
+                        content = delta.content
+                        if content is not None and content.strip():
+                            print(content, end='', flush=True)
+                            yield content
+        else:
+            # 非流式处理
+            if completion and completion.choices and len(completion.choices) > 0:
+                message = completion.choices[0].message
+                if message:
+                    content = message.content
+                    if content is not None and content.strip():
+                        yield content
+    except Exception as e:
+        print(f"大模型请求异常: {str(e)}", file=sys.stderr)
+        yield f"处理请求时发生异常: {str(e)}"
+
+# 同步获取大模型响应
+def get_xueban_response(query_text: str, stream: bool = True):
+    """
+    获取学伴角色的大模型响应
+    @param query_text: 查询文本
+    @param stream: 是否使用流式输出
+    @return: 完整响应文本
+    """
+    client = OpenAI(
+        api_key=LLM_API_KEY,
+        base_url=LLM_BASE_URL,
+    )
+    
+    # 创建请求
+    completion = client.chat.completions.create(
+        model=LLM_MODEL_NAME,
+        messages=[
+            {'role': 'system', 'content': prompt.strip()},
+            {'role': 'user', 'content': query_text}
+        ],
+        stream=stream
+    )
+    
+    full_response = []
+    
+    if stream:
+        for chunk in completion:
+            # 提取当前块的内容
+            if chunk.choices and chunk.choices[0].delta and chunk.choices[0].delta.content:
+                content = chunk.choices[0].delta.content
+                full_response.append(content)
+                # 实时输出内容，不换行
+                print(content, end='', flush=True)
+    else:
+        # 非流式处理
+        full_response.append(completion.choices[0].message.content)
+        
+    return ''.join(full_response)
+
+# 测试用例 main 函数
+def main():
+    """
+    测试学伴工具接口的主函数
+    """
+    print("===== 测试学伴工具接口 =====")
+    
+    # 测试同步接口
+    test_sync_interface()
+    
+    # 测试异步接口
+    import asyncio
+    print("\n测试异步接口...")
+    asyncio.run(test_async_interface())
+    
+    print("\n===== 测试完成 =====")
+
+def test_sync_interface():
+    """测试同步接口"""
+    print("\n测试同步接口...")
+    # 测试问题
+    questions = [
+        "你是谁？",
+        "讲个冷笑话",
+        "你男朋友是做什么的？"
+    ]
+    
+    for question in questions:
+        print(f"\n问题: {question}")
+        try:
+            # 调用同步接口获取响应
+            print("获取学伴响应中...")
+            response = get_xueban_response(question, stream=False)
+            print(f"学伴响应: {response}")
+            
+            # 简单验证响应
+            assert response.strip(), "响应内容为空"
+            print("✅ 同步接口测试通过")
+        except Exception as e:
+            print(f"❌ 同步接口测试失败: {str(e)}")
+
+async def test_async_interface():
+    """测试异步接口"""
+    # 测试问题
+    questions = [
+        "你是谁？",
+        "讲个冷笑话",
+        "你男朋友是做什么的？"
+    ]
+    
+    for question in questions:
+        print(f"\n问题: {question}")
+        try:
+            # 调用异步接口获取响应
+            print("获取学伴响应中...")
+            response_generator = get_xueban_response_async(question, stream=False)
+            response = ""
+            async for chunk in response_generator:
+                response += chunk
+            print(f"学伴响应: {response}")
+            
+            # 简单验证响应
+            assert response.strip(), "响应内容为空"
+            print("✅ 异步接口测试通过")
+        except Exception as e:
+            print(f"❌ 异步接口测试失败: {str(e)}")
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/dsLightRag/Util/__pycache__/XueBanUtil.cpython-310.pyc b/dsLightRag/Util/__pycache__/XueBanUtil.cpython-310.pyc
new file mode 100644
index 00000000..90aa8829
Binary files /dev/null and b/dsLightRag/Util/__pycache__/XueBanUtil.cpython-310.pyc differ