diff --git a/.dockerignore b/.dockerignore
index a3c2840..49a3e57 100644
--- a/.dockerignore
+++ b/.dockerignore
@@ -80,6 +80,10 @@ logs/
 *.log
 *.log.*
 
+# 数据临时文件
+data/temp/
+data/token.json
+
 # 数据库文件
 *.db
 *.sqlite
diff --git a/.github/workflows/docker.yml b/.github/workflows/docker.yml
index 74602dd..805a9e6 100644
--- a/.github/workflows/docker.yml
+++ b/.github/workflows/docker.yml
@@ -21,6 +21,17 @@ jobs:
       contents: read
       packages: write
       id-token: write
+    
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          # AMD64 平台
+          - platform: linux/amd64
+            suffix: amd64
+          # ARM64 平台
+          - platform: linux/arm64
+            suffix: arm64
 
     steps:
       - name: Checkout repository
@@ -42,19 +53,58 @@ jobs:
         with:
           images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
           tags: |
-            type=ref,event=branch
-            type=ref,event=pr
-            type=semver,pattern={{version}}
-            type=semver,pattern={{major}}.{{minor}}
-            type=raw,value=latest,enable={{is_default_branch}}
+            # 分支推送
+            type=ref,event=branch,suffix=-${{ matrix.suffix }}
+            # 语义化版本：完整版本号
+            type=semver,pattern={{version}},suffix=-${{ matrix.suffix }}
+            # latest标签
+            type=raw,value=latest-${{ matrix.suffix }},enable={{is_default_branch}}
 
       - name: Build and push Docker image
         uses: docker/build-push-action@v5
         with:
           context: .
-          platforms: linux/amd64,linux/arm64
+          platforms: ${{ matrix.platform }}
           push: ${{ github.event_name != 'pull_request' }}
           tags: ${{ steps.meta.outputs.tags }}
           labels: ${{ steps.meta.outputs.labels }}
-          cache-from: type=gha
-          cache-to: type=gha,mode=max
\ No newline at end of file
+          cache-from: type=gha,scope=${{ matrix.suffix }}
+          cache-to: type=gha,mode=max,scope=${{ matrix.suffix }}
+          pull: true
+  
+  # 合并多架构镜像为统一标签
+  merge-manifests:
+    runs-on: ubuntu-latest
+    needs: build-and-push
+    if: github.event_name != 'pull_request'
+    permissions:
+      contents: read
+      packages: write
+    
+    steps:
+      - name: Log in to Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+      
+      - name: Extract metadata
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+          tags: |
+            type=ref,event=branch
+            type=semver,pattern={{version}}
+            type=raw,value=latest,enable={{is_default_branch}}
+      
+      - name: Create and push manifest
+        run: |
+          TAGS="${{ steps.meta.outputs.tags }}"
+          for tag in $TAGS; do
+            echo "合并标签: $tag"
+            docker buildx imagetools create -t $tag \
+              ${tag}-amd64 \
+              ${tag}-arm64
+          done
\ No newline at end of file
diff --git a/Dockerfile b/Dockerfile
index 1992333..f625296 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -1,37 +1,55 @@
-FROM python:3.11-slim AS base
+# 构建阶段
+FROM python:3.11-slim AS builder
 
-WORKDIR /app
+WORKDIR /build
 
-# 安装依赖阶段
-FROM base AS dependencies
-
-# 安装运行时需要的系统依赖
-RUN apt-get update && \
-    apt-get install -y --no-install-recommends \
-    gcc \
-    && rm -rf /var/lib/apt/lists/*
-
-# 复制并安装 Python 依赖
+# 安装依赖到独立目录
 COPY requirements.txt .
-RUN pip install --no-cache-dir --upgrade pip && \
-    pip install --no-cache-dir -r requirements.txt
+RUN pip install --no-cache-dir --only-binary=:all: --prefix=/install -r requirements.txt && \
+  find /install -type d -name "__pycache__" -exec rm -rf {} + 2>/dev/null || true && \
+  find /install -type d -name "tests" -exec rm -rf {} + 2>/dev/null || true && \
+  find /install -type d -name "test" -exec rm -rf {} + 2>/dev/null || true && \
+  find /install -type d -name "*.dist-info" -exec sh -c 'rm -f "$1"/RECORD "$1"/INSTALLER' _ {} \; && \
+  find /install -type f -name "*.pyc" -delete && \
+  find /install -type f -name "*.pyo" -delete && \
+  find /install -name "*.so" -exec strip --strip-unneeded {} \; 2>/dev/null || true
 
-# 最终运行阶段
-FROM python:3.11-slim AS runtime
+# 运行阶段 - 使用最小镜像
+FROM python:3.11-slim
 
 WORKDIR /app
 
-# 复制必要的 Python 依赖
-COPY --from=dependencies /usr/local/lib/python3.11/site-packages /usr/local/lib/python3.11/site-packages
-COPY --from=dependencies /usr/local/bin /usr/local/bin
+# 清理基础镜像中的冗余文件
+RUN rm -rf /usr/share/doc/* \
+  /usr/share/man/* \
+  /usr/share/locale/* \
+  /var/cache/apt/* \
+  /var/lib/apt/lists/* \
+  /tmp/* \
+  /var/tmp/*
+
+# 从构建阶段复制已安装的包
+COPY --from=builder /install /usr/local
+
+# 创建必要的目录（包括用于挂载的data目录）
+RUN mkdir -p /app/logs /app/data/temp/image /app/data/temp/video
 
 # 复制应用代码
-COPY . .
+COPY app/ ./app/
+COPY main.py .
 
-# 创建必要的目录和文件
-RUN mkdir -p /app/logs /app/data/temp && \
-    echo '{"ssoNormal": {}, "ssoSuper": {}}' > /app/data/token.json
+# 复制并设置 entrypoint 脚本
+COPY docker-entrypoint.sh /usr/local/bin/
+RUN chmod +x /usr/local/bin/docker-entrypoint.sh
+
+# 删除 Python 字节码和缓存
+ENV PYTHONDONTWRITEBYTECODE=1 \
+  PYTHONUNBUFFERED=1
 
 EXPOSE 8000
 
-CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]
\ No newline at end of file
+# 使用 entrypoint 脚本初始化配置
+ENTRYPOINT ["docker-entrypoint.sh"]
+
+# 默认启动命令
+CMD ["python", "-m", "uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]
diff --git a/app/api/admin/manage.py b/app/api/admin/manage.py
index 3c8ca43..cb0fe81 100644
--- a/app/api/admin/manage.py
+++ b/app/api/admin/manage.py
@@ -1,8 +1,4 @@
-"""
-管理接口模块
-
-提供Token管理功能，包括登录验证、Token增删查等操作。
-"""
+"""管理接口 - Token管理和系统配置"""
 
 import secrets
 from typing import Dict, Any, List, Optional
@@ -18,10 +14,9 @@ from app.services.grok.token import token_manager
 from app.models.grok_models import TokenType
 
 
-# 创建路由器
 router = APIRouter(tags=["管理"])
 
-# 常量定义
+# 常量
 STATIC_DIR = Path(__file__).parents[2] / "template"
 TEMP_DIR = Path(__file__).parents[3] / "data" / "temp"
 IMAGE_CACHE_DIR = TEMP_DIR / "image"
@@ -30,68 +25,86 @@ SESSION_EXPIRE_HOURS = 24
 BYTES_PER_KB = 1024
 BYTES_PER_MB = 1024 * 1024
 
-# 简单的会话存储
+# 会话存储
 _sessions: Dict[str, datetime] = {}
 
 
 # === 请求/响应模型 ===
 
 class LoginRequest(BaseModel):
-    """登录请求"""
     username: str
     password: str
 
 
 class LoginResponse(BaseModel):
-    """登录响应"""
     success: bool
     token: Optional[str] = None
     message: str
 
 
 class AddTokensRequest(BaseModel):
-    """批量添加Token请求"""
     tokens: List[str]
-    token_type: str  # "sso" 或 "ssoSuper"
+    token_type: str
 
 
 class DeleteTokensRequest(BaseModel):
-    """批量删除Token请求"""
     tokens: List[str]
-    token_type: str  # "sso" 或 "ssoSuper"
+    token_type: str
 
 
 class TokenInfo(BaseModel):
-    """Token信息"""
     token: str
     token_type: str
     created_time: Optional[int] = None
     remaining_queries: int
     heavy_remaining_queries: int
-    status: str  # "未使用"、"限流中"、"失效"、"正常"
+    status: str
+    tags: List[str] = []
+    note: str = ""
 
 
 class TokenListResponse(BaseModel):
-    """Token列表响应"""
     success: bool
     data: List[TokenInfo]
     total: int
 
 
+class UpdateSettingsRequest(BaseModel):
+    global_config: Optional[Dict[str, Any]] = None
+    grok_config: Optional[Dict[str, Any]] = None
+
+
+class UpdateTokenTagsRequest(BaseModel):
+    token: str
+    token_type: str
+    tags: List[str]
+
+
+class UpdateTokenNoteRequest(BaseModel):
+    token: str
+    token_type: str
+    note: str
+
+
+class TestTokenRequest(BaseModel):
+    token: str
+    token_type: str
+
+
 # === 辅助函数 ===
 
 def validate_token_type(token_type_str: str) -> TokenType:
-    """验证并转换Token类型字符串为枚举"""
+    """验证Token类型"""
     if token_type_str not in ["sso", "ssoSuper"]:
         raise HTTPException(
             status_code=400,
-            detail={"error": "无效的Token类型，必须是 'sso' 或 'ssoSuper'", "code": "INVALID_TYPE"}
+            detail={"error": "无效的Token类型", "code": "INVALID_TYPE"}
         )
     return TokenType.NORMAL if token_type_str == "sso" else TokenType.SUPER
 
 
 def parse_created_time(created_time) -> Optional[int]:
-    """解析创建时间，统一处理不同格式"""
+    """解析创建时间"""
     if isinstance(created_time, str):
         return int(created_time) if created_time else None
     elif isinstance(created_time, int):
@@ -100,7 +113,7 @@ def parse_created_time(created_time) -> Optional[int]:
 
 
 def calculate_token_stats(tokens: Dict[str, Any], token_type: str) -> Dict[str, int]:
-    """计算Token统计信息"""
+    """计算Token统计"""
     total = len(tokens)
     expired = sum(1 for t in tokens.values() if t.get("status") == "expired")
 
@@ -111,7 +124,7 @@ def calculate_token_stats(tokens: Dict[str, Any], token_type: str) -> Dict[str,
                      if t.get("status") != "expired" and t.get("remainingQueries", -1) == 0)
         active = sum(1 for t in tokens.values()
                     if t.get("status") != "expired" and t.get("remainingQueries", -1) > 0)
-    else:  # super token
+    else:
         unused = sum(1 for t in tokens.values()
                     if t.get("status") != "expired" and
                     t.get("remainingQueries", -1) == -1 and t.get("heavyremainingQueries", -1) == -1)
@@ -122,69 +135,64 @@ def calculate_token_stats(tokens: Dict[str, Any], token_type: str) -> Dict[str,
                     if t.get("status") != "expired" and
                     (t.get("remainingQueries", -1) > 0 or t.get("heavyremainingQueries", -1) > 0))
 
-    return {
-        "total": total,
-        "unused": unused,
-        "limited": limited,
-        "expired": expired,
-        "active": active
-    }
+    return {"total": total, "unused": unused, "limited": limited, "expired": expired, "active": active}
 
 
 def verify_admin_session(authorization: Optional[str] = Header(None)) -> bool:
     """验证管理员会话"""
     if not authorization or not authorization.startswith("Bearer "):
-        raise HTTPException(
-            status_code=401,
-            detail={"error": "未授权访问", "code": "UNAUTHORIZED"}
-        )
+        raise HTTPException(status_code=401, detail={"error": "未授权访问", "code": "UNAUTHORIZED"})
     
-    token = authorization[7:]  # 移除 "Bearer " 前缀
+    token = authorization[7:]
     
-    # 检查token是否存在且未过期
     if token not in _sessions:
-        raise HTTPException(
-            status_code=401,
-            detail={"error": "会话已过期或无效", "code": "SESSION_INVALID"}
-        )
+        raise HTTPException(status_code=401, detail={"error": "会话无效", "code": "SESSION_INVALID"})
     
-    # 检查会话是否过期（24小时）
     if datetime.now() > _sessions[token]:
         del _sessions[token]
-        raise HTTPException(
-            status_code=401,
-            detail={"error": "会话已过期", "code": "SESSION_EXPIRED"}
-        )
+        raise HTTPException(status_code=401, detail={"error": "会话已过期", "code": "SESSION_EXPIRED"})
     
     return True
 
 
 def get_token_status(token_data: Dict[str, Any], token_type: str) -> str:
     """获取Token状态"""
-    # 首先检查是否失效（来自 token.json 的 status 字段）
     if token_data.get("status") == "expired":
         return "失效"
     
-    # 获取剩余次数
-    remaining_queries = token_data.get("remainingQueries", -1)
+    remaining = token_data.get("remainingQueries", -1)
     heavy_remaining = token_data.get("heavyremainingQueries", -1)
     
-    # 根据token类型选择正确的字段
-    if token_type == "ssoSuper":
-        # Super token 可能使用 heavy 模型
-        relevant_remaining = max(remaining_queries, heavy_remaining)
-    else:
-        # 普通token主要看 remaining_queries
-        relevant_remaining = remaining_queries
+    relevant = max(remaining, heavy_remaining) if token_type == "ssoSuper" else remaining
     
-    if relevant_remaining == -1:
+    if relevant == -1:
         return "未使用"
-    elif relevant_remaining == 0:
+    elif relevant == 0:
         return "限流中"
     else:
         return "正常"
 
 
+def _calculate_dir_size(directory: Path) -> int:
+    """计算目录大小"""
+    total = 0
+    for file_path in directory.iterdir():
+        if file_path.is_file():
+            try:
+                total += file_path.stat().st_size
+            except Exception as e:
+                logger.warning(f"[Admin] 无法获取文件大小: {file_path.name}, {e}")
+    return total
+
+
+def _format_size(size_bytes: int) -> str:
+    """格式化文件大小"""
+    size_mb = size_bytes / BYTES_PER_MB
+    if size_mb < 1:
+        return f"{size_bytes / BYTES_PER_KB:.1f} KB"
+    return f"{size_mb:.1f} MB"
+
+
 # === 页面路由 ===
 
 @router.get("/login", response_class=HTMLResponse)
@@ -209,292 +217,163 @@ async def manage_page():
 
 @router.post("/api/login", response_model=LoginResponse)
 async def admin_login(request: LoginRequest) -> LoginResponse:
-    """
-    管理员登录
-    
-    验证用户名和密码，成功后返回会话token。
-    """
+    """管理员登录"""
     try:
-        logger.debug(f"[Admin] 管理员登录尝试 - 用户名: {request.username}")
+        logger.debug(f"[Admin] 登录尝试: {request.username}")
 
-        # 验证用户名和密码
-        expected_username = setting.global_config.get("admin_username", "")
-        expected_password = setting.global_config.get("admin_password", "")
+        expected_user = setting.global_config.get("admin_username", "")
+        expected_pass = setting.global_config.get("admin_password", "")
 
-        if request.username != expected_username or request.password != expected_password:
-            logger.warning(f"[Admin] 登录失败: 用户名或密码错误 - 用户名: {request.username}")
-            return LoginResponse(
-                success=False,
-                message="用户名或密码错误"
-            )
+        if request.username != expected_user or request.password != expected_pass:
+            logger.warning(f"[Admin] 登录失败: {request.username}")
+            return LoginResponse(success=False, message="用户名或密码错误")
 
-        # 生成会话token
         session_token = secrets.token_urlsafe(32)
+        _sessions[session_token] = datetime.now() + timedelta(hours=SESSION_EXPIRE_HOURS)
 
-        # 设置会话过期时间
-        expire_time = datetime.now() + timedelta(hours=SESSION_EXPIRE_HOURS)
-        _sessions[session_token] = expire_time
-
-        logger.debug(f"[Admin] 管理员登录成功 - 用户名: {request.username}")
-
-        return LoginResponse(
-            success=True,
-            token=session_token,
-            message="登录成功"
-        )
+        logger.debug(f"[Admin] 登录成功: {request.username}")
+        return LoginResponse(success=True, token=session_token, message="登录成功")
 
     except Exception as e:
-        logger.error(f"[Admin] 登录处理异常 - 用户名: {request.username}, 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"登录失败: {str(e)}", "code": "LOGIN_ERROR"}
-        )
+        logger.error(f"[Admin] 登录异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"登录失败: {e}", "code": "LOGIN_ERROR"})
 
 
 @router.post("/api/logout")
-async def admin_logout(_: bool = Depends(verify_admin_session),
-                       authorization: Optional[str] = Header(None)) -> Dict[str, Any]:
-    """
-    管理员登出
-    
-    清除会话token。
-    """
+async def admin_logout(_: bool = Depends(verify_admin_session), authorization: Optional[str] = Header(None)) -> Dict[str, Any]:
+    """管理员登出"""
     try:
         if authorization and authorization.startswith("Bearer "):
             token = authorization[7:]
             if token in _sessions:
                 del _sessions[token]
-                logger.debug("[Admin] 管理员登出成功")
+                logger.debug("[Admin] 登出成功")
                 return {"success": True, "message": "登出成功"}
 
-        logger.warning("[Admin] 登出失败: 无效或缺失的会话Token")
+        logger.warning("[Admin] 登出失败: 无效会话")
         return {"success": False, "message": "无效的会话"}
 
     except Exception as e:
-        logger.error(f"[Admin] 登出处理异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"登出失败: {str(e)}", "code": "LOGOUT_ERROR"}
-        )
+        logger.error(f"[Admin] 登出异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"登出失败: {e}", "code": "LOGOUT_ERROR"})
 
 
 @router.get("/api/tokens", response_model=TokenListResponse)
 async def list_tokens(_: bool = Depends(verify_admin_session)) -> TokenListResponse:
-    """
-    获取所有Token列表
-    
-    返回系统中所有Token及其状态信息。
-    """
+    """获取Token列表"""
     try:
-        logger.debug("[Admin] 开始获取Token列表")
+        logger.debug("[Admin] 获取Token列表")
 
-        all_tokens_data = token_manager.get_tokens()
+        all_tokens = token_manager.get_tokens()
         token_list: List[TokenInfo] = []
 
-        # 处理普通Token
-        normal_tokens = all_tokens_data.get(TokenType.NORMAL.value, {})
-        for token, data in normal_tokens.items():
+        # 普通Token
+        for token, data in all_tokens.get(TokenType.NORMAL.value, {}).items():
             token_list.append(TokenInfo(
                 token=token,
                 token_type="sso",
                 created_time=parse_created_time(data.get("createdTime")),
                 remaining_queries=data.get("remainingQueries", -1),
                 heavy_remaining_queries=data.get("heavyremainingQueries", -1),
-                status=get_token_status(data, "sso")
+                status=get_token_status(data, "sso"),
+                tags=data.get("tags", []),
+                note=data.get("note", "")
             ))
 
-        # 处理Super Token
-        super_tokens = all_tokens_data.get(TokenType.SUPER.value, {})
-        for token, data in super_tokens.items():
+        # Super Token
+        for token, data in all_tokens.get(TokenType.SUPER.value, {}).items():
             token_list.append(TokenInfo(
                 token=token,
                 token_type="ssoSuper",
                 created_time=parse_created_time(data.get("createdTime")),
                 remaining_queries=data.get("remainingQueries", -1),
                 heavy_remaining_queries=data.get("heavyremainingQueries", -1),
-                status=get_token_status(data, "ssoSuper")
+                status=get_token_status(data, "ssoSuper"),
+                tags=data.get("tags", []),
+                note=data.get("note", "")
             ))
 
-        normal_count = len(normal_tokens)
-        super_count = len(super_tokens)
-        total_count = len(token_list)
-
-        logger.debug(f"[Admin] Token列表获取成功 - 普通Token: {normal_count}, Super Token: {super_count}, 总计: {total_count}")
-
-        return TokenListResponse(
-            success=True,
-            data=token_list,
-            total=total_count
-        )
+        logger.debug(f"[Admin] Token列表获取成功: {len(token_list)}个")
+        return TokenListResponse(success=True, data=token_list, total=len(token_list))
 
     except Exception as e:
-        logger.error(f"[Admin] 获取Token列表异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"获取Token列表失败: {str(e)}", "code": "LIST_ERROR"}
-        )
+        logger.error(f"[Admin] 获取Token列表异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"获取失败: {e}", "code": "LIST_ERROR"})
 
 
 @router.post("/api/tokens/add")
-async def add_tokens(request: AddTokensRequest,
-                    _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """
-    批量添加Token
-    
-    支持添加普通Token(sso)和Super Token(ssoSuper)。
-    """
+async def add_tokens(request: AddTokensRequest, _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
+    """批量添加Token"""
     try:
-        logger.debug(f"[Admin] 批量添加Token - 类型: {request.token_type}, 数量: {len(request.tokens)}")
+        logger.debug(f"[Admin] 添加Token: {request.token_type}, {len(request.tokens)}个")
 
-        # 验证并转换token类型
         token_type = validate_token_type(request.token_type)
-
-        # 添加Token
         await token_manager.add_token(request.tokens, token_type)
 
-        logger.debug(f"[Admin] Token添加成功 - 类型: {request.token_type}, 数量: {len(request.tokens)}")
-
-        return {
-            "success": True,
-            "message": f"成功添加 {len(request.tokens)} 个Token",
-            "count": len(request.tokens)
-        }
+        logger.debug(f"[Admin] Token添加成功: {len(request.tokens)}个")
+        return {"success": True, "message": f"成功添加 {len(request.tokens)} 个Token", "count": len(request.tokens)}
 
     except HTTPException:
         raise
     except Exception as e:
-        logger.error(f"[Admin] Token添加异常 - 类型: {request.token_type}, 数量: {len(request.tokens)}, 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"添加Token失败: {str(e)}", "code": "ADD_ERROR"}
-        )
+        logger.error(f"[Admin] Token添加异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"添加失败: {e}", "code": "ADD_ERROR"})
 
 
 @router.post("/api/tokens/delete")
-async def delete_tokens(request: DeleteTokensRequest,
-                       _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """
-    批量删除Token
-    
-    支持删除普通Token(sso)和Super Token(ssoSuper)。
-    """
+async def delete_tokens(request: DeleteTokensRequest, _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
+    """批量删除Token"""
     try:
-        logger.debug(f"[Admin] 批量删除Token - 类型: {request.token_type}, 数量: {len(request.tokens)}")
+        logger.debug(f"[Admin] 删除Token: {request.token_type}, {len(request.tokens)}个")
 
-        # 验证并转换token类型
         token_type = validate_token_type(request.token_type)
-
-        # 删除Token
         await token_manager.delete_token(request.tokens, token_type)
 
-        logger.debug(f"[Admin] Token删除成功 - 类型: {request.token_type}, 数量: {len(request.tokens)}")
-
-        return {
-            "success": True,
-            "message": f"成功删除 {len(request.tokens)} 个Token",
-            "count": len(request.tokens)
-        }
+        logger.debug(f"[Admin] Token删除成功: {len(request.tokens)}个")
+        return {"success": True, "message": f"成功删除 {len(request.tokens)} 个Token", "count": len(request.tokens)}
 
     except HTTPException:
         raise
     except Exception as e:
-        logger.error(f"[Admin] Token删除异常 - 类型: {request.token_type}, 数量: {len(request.tokens)}, 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"删除Token失败: {str(e)}", "code": "DELETE_ERROR"}
-        )
+        logger.error(f"[Admin] Token删除异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"删除失败: {e}", "code": "DELETE_ERROR"})
 
 
 @router.get("/api/settings")
 async def get_settings(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """获取全局配置"""
+    """获取配置"""
     try:
-        logger.debug("[Admin] 获取全局配置")
-        return {
-            "success": True,
-            "data": {
-                "global": setting.global_config,
-                "grok": setting.grok_config
-            }
-        }
+        logger.debug("[Admin] 获取配置")
+        return {"success": True, "data": {"global": setting.global_config, "grok": setting.grok_config}}
     except Exception as e:
-        logger.error(f"[Admin] 获取配置失败: {str(e)}")
-        raise HTTPException(status_code=500, detail={"error": f"获取配置失败: {str(e)}", "code": "GET_SETTINGS_ERROR"})
-
-
-class UpdateSettingsRequest(BaseModel):
-    """更新配置请求"""
-    global_config: Optional[Dict[str, Any]] = None
-    grok_config: Optional[Dict[str, Any]] = None
-
-
-class StreamTimeoutSettings(BaseModel):
-    """流式超时配置"""
-    stream_chunk_timeout: int = 120
-    stream_first_response_timeout: int = 30
-    stream_total_timeout: int = 600
+        logger.error(f"[Admin] 获取配置失败: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"获取失败: {e}", "code": "GET_SETTINGS_ERROR"})
 
 
 @router.post("/api/settings")
 async def update_settings(request: UpdateSettingsRequest, _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """更新全局配置"""
+    """更新配置"""
     try:
-        logger.debug("[Admin] 更新全局配置")
-
-        # 使用ConfigManager的save方法（支持存储抽象层）
-        await setting.save(
-            global_config=request.global_config,
-            grok_config=request.grok_config
-        )
-
+        logger.debug("[Admin] 更新配置")
+        await setting.save(global_config=request.global_config, grok_config=request.grok_config)
         logger.debug("[Admin] 配置更新成功")
         return {"success": True, "message": "配置更新成功"}
     except Exception as e:
-        logger.error(f"[Admin] 更新配置失败: {str(e)}")
-        raise HTTPException(status_code=500, detail={"error": f"更新配置失败: {str(e)}", "code": "UPDATE_SETTINGS_ERROR"})
-
-
-def _calculate_dir_size(directory: Path) -> int:
-    """计算目录中所有文件的大小（字节）"""
-    total_size = 0
-    for file_path in directory.iterdir():
-        if file_path.is_file():
-            try:
-                total_size += file_path.stat().st_size
-            except Exception as e:
-                logger.warning(f"[Admin] 无法获取文件大小: {file_path.name}, 错误: {str(e)}")
-    return total_size
-
-
-def _format_size(size_bytes: int) -> str:
-    """格式化字节大小为可读字符串"""
-    size_mb = size_bytes / BYTES_PER_MB
-    if size_mb < 1:
-        size_kb = size_bytes / BYTES_PER_KB
-        return f"{size_kb:.1f} KB"
-    return f"{size_mb:.1f} MB"
+        logger.error(f"[Admin] 更新配置失败: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"更新失败: {e}", "code": "UPDATE_SETTINGS_ERROR"})
 
 
 @router.get("/api/cache/size")
 async def get_cache_size(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
     """获取缓存大小"""
     try:
-        logger.debug("[Admin] 开始获取缓存大小")
+        logger.debug("[Admin] 获取缓存大小")
 
-        # 计算图片缓存大小
-        image_size = 0
-        if IMAGE_CACHE_DIR.exists():
-            image_size = _calculate_dir_size(IMAGE_CACHE_DIR)
-        
-        # 计算视频缓存大小
-        video_size = 0
-        if VIDEO_CACHE_DIR.exists():
-            video_size = _calculate_dir_size(VIDEO_CACHE_DIR)
-        
-        # 总大小
+        image_size = _calculate_dir_size(IMAGE_CACHE_DIR) if IMAGE_CACHE_DIR.exists() else 0
+        video_size = _calculate_dir_size(VIDEO_CACHE_DIR) if VIDEO_CACHE_DIR.exists() else 0
         total_size = image_size + video_size
 
-        logger.debug(f"[Admin] 缓存大小获取完成 - 图片: {_format_size(image_size)}, 视频: {_format_size(video_size)}, 总计: {_format_size(total_size)}")
+        logger.debug(f"[Admin] 缓存大小: 图片{_format_size(image_size)}, 视频{_format_size(video_size)}")
         
         return {
             "success": True,
@@ -509,213 +388,235 @@ async def get_cache_size(_: bool = Depends(verify_admin_session)) -> Dict[str, A
         }
 
     except Exception as e:
-        logger.error(f"[Admin] 获取缓存大小异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"获取缓存大小失败: {str(e)}", "code": "CACHE_SIZE_ERROR"}
-        )
+        logger.error(f"[Admin] 获取缓存大小异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"获取失败: {e}", "code": "CACHE_SIZE_ERROR"})
 
 
 @router.post("/api/cache/clear")
 async def clear_cache(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """清理缓存
-
-    删除所有临时文件"""
+    """清理所有缓存"""
     try:
-        logger.debug("[Admin] 开始清理缓存")
+        logger.debug("[Admin] 清理缓存")
 
-        deleted_count = 0
         image_count = 0
         video_count = 0
 
-        # 清理图片缓存
+        # 清理图片
         if IMAGE_CACHE_DIR.exists():
             for file_path in IMAGE_CACHE_DIR.iterdir():
                 if file_path.is_file():
                     try:
                         file_path.unlink()
                         image_count += 1
-                        logger.debug(f"[Admin] 删除图片缓存: {file_path.name}")
                     except Exception as e:
-                        logger.error(f"[Admin] 删除图片缓存失败: {file_path.name}, 错误: {str(e)}")
+                        logger.error(f"[Admin] 删除失败: {file_path.name}, {e}")
 
-        # 清理视频缓存
+        # 清理视频
         if VIDEO_CACHE_DIR.exists():
             for file_path in VIDEO_CACHE_DIR.iterdir():
                 if file_path.is_file():
                     try:
                         file_path.unlink()
                         video_count += 1
-                        logger.debug(f"[Admin] 删除视频缓存: {file_path.name}")
                     except Exception as e:
-                        logger.error(f"[Admin] 删除视频缓存失败: {file_path.name}, 错误: {str(e)}")
+                        logger.error(f"[Admin] 删除失败: {file_path.name}, {e}")
 
-        deleted_count = image_count + video_count
-        logger.debug(f"[Admin] 缓存清理完成 - 图片: {image_count}, 视频: {video_count}, 总计: {deleted_count}")
+        total = image_count + video_count
+        logger.debug(f"[Admin] 缓存清理完成: 图片{image_count}, 视频{video_count}")
 
         return {
             "success": True,
-            "message": f"成功清理缓存，删除图片 {image_count} 个，视频 {video_count} 个，共 {deleted_count} 个文件",
-            "data": {
-                "deleted_count": deleted_count,
-                "image_count": image_count,
-                "video_count": video_count
-            }
+            "message": f"成功清理缓存，删除图片 {image_count} 个，视频 {video_count} 个，共 {total} 个文件",
+            "data": {"deleted_count": total, "image_count": image_count, "video_count": video_count}
         }
 
     except Exception as e:
-        logger.error(f"[Admin] 清理缓存异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"清理缓存失败: {str(e)}", "code": "CACHE_CLEAR_ERROR"}
-        )
+        logger.error(f"[Admin] 清理缓存异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"清理失败: {e}", "code": "CACHE_CLEAR_ERROR"})
 
 
 @router.post("/api/cache/clear/images")
 async def clear_image_cache(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """清理图片缓存
-
-    仅删除图片缓存文件"""
+    """清理图片缓存"""
     try:
-        logger.debug("[Admin] 开始清理图片缓存")
+        logger.debug("[Admin] 清理图片缓存")
 
-        deleted_count = 0
-
-        # 清理图片缓存
+        count = 0
         if IMAGE_CACHE_DIR.exists():
             for file_path in IMAGE_CACHE_DIR.iterdir():
                 if file_path.is_file():
                     try:
                         file_path.unlink()
-                        deleted_count += 1
-                        logger.debug(f"[Admin] 删除图片缓存: {file_path.name}")
+                        count += 1
                     except Exception as e:
-                        logger.error(f"[Admin] 删除图片缓存失败: {file_path.name}, 错误: {str(e)}")
+                        logger.error(f"[Admin] 删除失败: {file_path.name}, {e}")
 
-        logger.debug(f"[Admin] 图片缓存清理完成 - 删除 {deleted_count} 个文件")
-
-        return {
-            "success": True,
-            "message": f"成功清理图片缓存，删除 {deleted_count} 个文件",
-            "data": {
-                "deleted_count": deleted_count,
-                "type": "images"
-            }
-        }
+        logger.debug(f"[Admin] 图片缓存清理完成: {count}个")
+        return {"success": True, "message": f"成功清理图片缓存，删除 {count} 个文件", "data": {"deleted_count": count, "type": "images"}}
 
     except Exception as e:
-        logger.error(f"[Admin] 清理图片缓存异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"清理图片缓存失败: {str(e)}", "code": "IMAGE_CACHE_CLEAR_ERROR"}
-        )
+        logger.error(f"[Admin] 清理图片缓存异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"清理失败: {e}", "code": "IMAGE_CACHE_CLEAR_ERROR"})
 
 
 @router.post("/api/cache/clear/videos")
 async def clear_video_cache(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """清理视频缓存
-
-    仅删除视频缓存文件"""
+    """清理视频缓存"""
     try:
-        logger.debug("[Admin] 开始清理视频缓存")
+        logger.debug("[Admin] 清理视频缓存")
 
-        deleted_count = 0
-
-        # 清理视频缓存
+        count = 0
         if VIDEO_CACHE_DIR.exists():
             for file_path in VIDEO_CACHE_DIR.iterdir():
                 if file_path.is_file():
                     try:
                         file_path.unlink()
-                        deleted_count += 1
-                        logger.debug(f"[Admin] 删除视频缓存: {file_path.name}")
+                        count += 1
                     except Exception as e:
-                        logger.error(f"[Admin] 删除视频缓存失败: {file_path.name}, 错误: {str(e)}")
+                        logger.error(f"[Admin] 删除失败: {file_path.name}, {e}")
 
-        logger.debug(f"[Admin] 视频缓存清理完成 - 删除 {deleted_count} 个文件")
-
-        return {
-            "success": True,
-            "message": f"成功清理视频缓存，删除 {deleted_count} 个文件",
-            "data": {
-                "deleted_count": deleted_count,
-                "type": "videos"
-            }
-        }
+        logger.debug(f"[Admin] 视频缓存清理完成: {count}个")
+        return {"success": True, "message": f"成功清理视频缓存，删除 {count} 个文件", "data": {"deleted_count": count, "type": "videos"}}
 
     except Exception as e:
-        logger.error(f"[Admin] 清理视频缓存异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"清理视频缓存失败: {str(e)}", "code": "VIDEO_CACHE_CLEAR_ERROR"}
-        )
+        logger.error(f"[Admin] 清理视频缓存异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"清理失败: {e}", "code": "VIDEO_CACHE_CLEAR_ERROR"})
 
 
 @router.get("/api/stats")
 async def get_stats(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """
-    获取统计信息
-
-    返回Token的统计数据。
-    """
+    """获取统计信息"""
     try:
         logger.debug("[Admin] 开始获取统计信息")
 
-        all_tokens_data = token_manager.get_tokens()
+        all_tokens = token_manager.get_tokens()
+        normal_stats = calculate_token_stats(all_tokens.get(TokenType.NORMAL.value, {}), "normal")
+        super_stats = calculate_token_stats(all_tokens.get(TokenType.SUPER.value, {}), "super")
+        total = normal_stats["total"] + super_stats["total"]
 
-        # 统计普通Token
-        normal_tokens = all_tokens_data.get(TokenType.NORMAL.value, {})
-        normal_stats = calculate_token_stats(normal_tokens, "normal")
-
-        # 统计Super Token
-        super_tokens = all_tokens_data.get(TokenType.SUPER.value, {})
-        super_stats = calculate_token_stats(super_tokens, "super")
-
-        total_count = normal_stats["total"] + super_stats["total"]
-
-        stats = {
-            "success": True,
-            "data": {
-                "normal": normal_stats,
-                "super": super_stats,
-                "total": total_count
-            }
-        }
-
-        logger.debug(f"[Admin] 统计信息获取成功 - 普通Token: {normal_stats['total']}, Super Token: {super_stats['total']}, 总计: {total_count}")
-        return stats
+        logger.debug(f"[Admin] 统计信息获取成功 - 普通Token: {normal_stats['total']}, Super Token: {super_stats['total']}, 总计: {total}")
+        return {"success": True, "data": {"normal": normal_stats, "super": super_stats, "total": total}}
 
     except Exception as e:
-        logger.error(f"[Admin] 获取统计信息异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"获取统计信息失败: {str(e)}", "code": "STATS_ERROR"}
-        )
+        logger.error(f"[Admin] 获取统计信息异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"获取失败: {e}", "code": "STATS_ERROR"})
 
 
 @router.get("/api/storage/mode")
 async def get_storage_mode(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
-    """
-    获取当前存储模式
-
-    返回当前的存储模式（file/mysql/redis）。
-    """
+    """获取存储模式"""
     try:
         logger.debug("[Admin] 获取存储模式")
-
         import os
-        storage_mode = os.getenv("STORAGE_MODE", "file").upper()
+        mode = os.getenv("STORAGE_MODE", "file").upper()
+        return {"success": True, "data": {"mode": mode}}
+    except Exception as e:
+        logger.error(f"[Admin] 获取存储模式异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"获取失败: {e}", "code": "STORAGE_MODE_ERROR"})
 
-        return {
-            "success": True,
-            "data": {
-                "mode": storage_mode
-            }
-        }
+
+@router.post("/api/tokens/tags")
+async def update_token_tags(request: UpdateTokenTagsRequest, _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
+    """更新Token标签"""
+    try:
+        logger.debug(f"[Admin] 更新Token标签: {request.token[:10]}..., {request.tags}")
+
+        token_type = validate_token_type(request.token_type)
+        await token_manager.update_token_tags(request.token, token_type, request.tags)
+
+        logger.debug(f"[Admin] Token标签更新成功: {request.token[:10]}...")
+        return {"success": True, "message": "标签更新成功", "tags": request.tags}
+
+    except HTTPException:
+        raise
+    except Exception as e:
+        logger.error(f"[Admin] Token标签更新异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"更新失败: {e}", "code": "UPDATE_TAGS_ERROR"})
+
+
+@router.get("/api/tokens/tags/all")
+async def get_all_tags(_: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
+    """获取所有标签"""
+    try:
+        logger.debug("[Admin] 获取所有标签")
+
+        all_tokens = token_manager.get_tokens()
+        tags_set = set()
+
+        for token_type_data in all_tokens.values():
+            for token_data in token_type_data.values():
+                tags = token_data.get("tags", [])
+                if isinstance(tags, list):
+                    tags_set.update(tags)
+
+        tags_list = sorted(list(tags_set))
+        logger.debug(f"[Admin] 标签获取成功: {len(tags_list)}个")
+        return {"success": True, "data": tags_list}
 
     except Exception as e:
-        logger.error(f"[Admin] 获取存储模式异常 - 错误: {str(e)}")
-        raise HTTPException(
-            status_code=500,
-            detail={"error": f"获取存储模式失败: {str(e)}", "code": "STORAGE_MODE_ERROR"}
-        )
+        logger.error(f"[Admin] 获取标签异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"获取失败: {e}", "code": "GET_TAGS_ERROR"})
+
+
+@router.post("/api/tokens/note")
+async def update_token_note(request: UpdateTokenNoteRequest, _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
+    """更新Token备注"""
+    try:
+        logger.debug(f"[Admin] 更新Token备注: {request.token[:10]}...")
+
+        token_type = validate_token_type(request.token_type)
+        await token_manager.update_token_note(request.token, token_type, request.note)
+
+        logger.debug(f"[Admin] Token备注更新成功: {request.token[:10]}...")
+        return {"success": True, "message": "备注更新成功", "note": request.note}
+
+    except HTTPException:
+        raise
+    except Exception as e:
+        logger.error(f"[Admin] Token备注更新异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"更新失败: {e}", "code": "UPDATE_NOTE_ERROR"})
+
+
+@router.post("/api/tokens/test")
+async def test_token(request: TestTokenRequest, _: bool = Depends(verify_admin_session)) -> Dict[str, Any]:
+    """测试Token可用性"""
+    try:
+        logger.debug(f"[Admin] 测试Token: {request.token[:10]}...")
+
+        token_type = validate_token_type(request.token_type)
+        auth_token = f"sso-rw={request.token};sso={request.token}"
+
+        result = await token_manager.check_limits(auth_token, "grok-4-fast")
+
+        if result:
+            logger.debug(f"[Admin] Token测试成功: {request.token[:10]}...")
+            return {
+                "success": True,
+                "message": "Token有效",
+                "data": {
+                    "valid": True,
+                    "remaining_queries": result.get("remainingTokens", -1),
+                    "limit": result.get("limit", -1)
+                }
+            }
+        else:
+            logger.warning(f"[Admin] Token测试失败: {request.token[:10]}...")
+
+            all_tokens = token_manager.get_tokens()
+            token_data = all_tokens.get(token_type.value, {}).get(request.token)
+
+            if token_data:
+                if token_data.get("status") == "expired":
+                    return {"success": False, "message": "Token已失效", "data": {"valid": False, "error_type": "expired", "error_code": 401}}
+                elif token_data.get("remainingQueries") == 0:
+                    return {"success": False, "message": "Token已被限流", "data": {"valid": False, "error_type": "limited", "error_code": "other"}}
+                else:
+                    return {"success": False, "message": "服务器被block或网络错误", "data": {"valid": False, "error_type": "blocked", "error_code": 403}}
+            else:
+                return {"success": False, "message": "Token数据异常", "data": {"valid": False, "error_type": "unknown", "error_code": "data_error"}}
+
+    except HTTPException:
+        raise
+    except Exception as e:
+        logger.error(f"[Admin] Token测试异常: {e}")
+        raise HTTPException(status_code=500, detail={"error": f"测试失败: {e}", "code": "TEST_TOKEN_ERROR"})
diff --git a/app/api/v1/chat.py b/app/api/v1/chat.py
index 87460e5..d80424d 100644
--- a/app/api/v1/chat.py
+++ b/app/api/v1/chat.py
@@ -1,9 +1,4 @@
-# -*- coding: utf-8 -*-
-"""
-聊天API路由模块
-
-提供OpenAI兼容的聊天API接口，支持与Grok模型的交互。
-"""
+"""聊天API路由 - OpenAI兼容的聊天接口"""
 
 from fastapi import APIRouter, Depends, HTTPException
 from typing import Optional
@@ -15,38 +10,20 @@ from app.core.logger import logger
 from app.services.grok.client import GrokClient
 from app.models.openai_schema import OpenAIChatRequest
 
-# 聊天路由
+
 router = APIRouter(prefix="/chat", tags=["聊天"])
 
 
 @router.post("/completions", response_model=None)
-async def chat_completions(
-    request: OpenAIChatRequest,
-    _: Optional[str] = Depends(auth_manager.verify)
-):
-    """
-    创建聊天补全
-
-    兼容OpenAI聊天API的端点，支持流式和非流式响应。
-
-    Args:
-        request: OpenAI格式的聊天请求
-        _: 认证依赖（自动验证）
-
-    Returns:
-        OpenAIChatCompletionResponse: 非流式响应
-        StreamingResponse: 流式响应
-
-    Raises:
-        HTTPException: 当请求处理失败时
-    """
+async def chat_completions(request: OpenAIChatRequest, _: Optional[str] = Depends(auth_manager.verify)):
+    """创建聊天补全（支持流式和非流式）"""
     try:
-        logger.info(f"[Chat] 聊天请求 - 模型: {request.model}")
+        logger.info("[Chat] 收到聊天请求")
 
-        # 调用Grok客户端处理请求
+        # 调用Grok客户端
         result = await GrokClient.openai_to_grok(request.model_dump())
         
-        # 如果是流式响应，GrokClient已经返回了Iterator，直接包装为StreamingResponse
+        # 流式响应
         if request.stream:
             return StreamingResponse(
                 content=result,
@@ -58,11 +35,11 @@ async def chat_completions(
                 }
             )
         
-        # 非流式响应直接返回
+        # 非流式响应
         return result
         
     except GrokApiException as e:
-        logger.error(f"[Chat] Grok API错误: {str(e)} - 详情: {e.details}")
+        logger.error(f"[Chat] Grok API错误: {e} - 详情: {e.details}")
         raise HTTPException(
             status_code=500,
             detail={
@@ -74,7 +51,7 @@ async def chat_completions(
             }
         )
     except Exception as e:
-        logger.error(f"[Chat] 聊天请求处理失败: {str(e)}")
+        logger.error(f"[Chat] 处理失败: {e}")
         raise HTTPException(
             status_code=500,
             detail={
diff --git a/app/api/v1/images.py b/app/api/v1/images.py
index b9c191e..e0679fa 100644
--- a/app/api/v1/images.py
+++ b/app/api/v1/images.py
@@ -1,4 +1,4 @@
-"""图片服务API路由"""
+"""图片服务API - 提供缓存的图片和视频文件"""
 
 from fastapi import APIRouter, HTTPException
 from fastapi.responses import FileResponse
@@ -13,31 +13,26 @@ router = APIRouter()
 @router.get("/images/{img_path:path}")
 async def get_image(img_path: str):
     """获取缓存的图片或视频
-
+    
     Args:
-        img_path: 文件路径，格式如 users-xxx-generated-xxx-image.jpg 或 users-xxx-generated-xxx-video.mp4
-
-    Returns:
-        文件响应
+        img_path: 文件路径（格式：users-xxx-generated-xxx-image.jpg）
     """
     try:
-        # 将路径转换回原始格式（短横线转斜杠）
+        # 转换路径（短横线→斜杠）
         original_path = "/" + img_path.replace('-', '/')
 
-        # 判断是图片还是视频
+        # 判断类型
         is_video = any(original_path.lower().endswith(ext) for ext in ['.mp4', '.webm', '.mov', '.avi'])
         
         if is_video:
-            # 检查视频缓存
             cache_path = video_cache_service.get_cached(original_path)
             media_type = "video/mp4"
         else:
-            # 检查图片缓存
             cache_path = image_cache_service.get_cached(original_path)
             media_type = "image/jpeg"
 
         if cache_path and cache_path.exists():
-            logger.debug(f"[MediaAPI] 返回缓存文件: {cache_path}")
+            logger.debug(f"[MediaAPI] 返回缓存: {cache_path}")
             return FileResponse(
                 path=str(cache_path),
                 media_type=media_type,
@@ -48,11 +43,11 @@ async def get_image(img_path: str):
             )
 
         # 文件不存在
-        logger.warning(f"[MediaAPI] 文件未找到: {original_path}")
+        logger.warning(f"[MediaAPI] 未找到: {original_path}")
         raise HTTPException(status_code=404, detail="File not found")
 
     except HTTPException:
         raise
     except Exception as e:
-        logger.error(f"[MediaAPI] 获取文件失败: {e}")
+        logger.error(f"[MediaAPI] 获取失败: {e}")
         raise HTTPException(status_code=500, detail=str(e))
diff --git a/app/api/v1/models.py b/app/api/v1/models.py
index 5c18513..5cdcc71 100644
--- a/app/api/v1/models.py
+++ b/app/api/v1/models.py
@@ -1,8 +1,4 @@
-"""
-模型接口模块
-
-提供 OpenAI 兼容的 /v1/models 端点，返回系统支持的所有模型列表。
-"""
+"""模型接口 - OpenAI兼容的模型列表端点"""
 
 import time
 from typing import Dict, Any, List, Optional
@@ -12,43 +8,27 @@ from app.models.grok_models import Models
 from app.core.auth import auth_manager
 from app.core.logger import logger
 
-# 配置日志
 
-# 创建路由器
 router = APIRouter(tags=["模型"])
 
 
 @router.get("/models")
 async def list_models(_: Optional[str] = Depends(auth_manager.verify)) -> Dict[str, Any]:
-    """
-    获取可用模型列表
-
-    返回 OpenAI 兼容的模型列表格式，包含系统支持的所有 Grok 模型的详细信息。
-
-    Args:
-        _: 认证依赖（自动验证）
-
-    Returns:
-        Dict[str, Any]: 包含模型列表的响应数据
-    """
+    """获取可用模型列表"""
     try:
-        logger.debug("[Models] 请求获取模型列表")
+        logger.debug("[Models] 请求模型列表")
 
-        # 获取当前时间戳
-        current_timestamp = int(time.time())
-        
-        # 构建模型数据列表
+        timestamp = int(time.time())
         model_data: List[Dict[str, Any]] = []
         
         for model in Models:
             model_id = model.value
             config = Models.get_model_info(model_id)
             
-            # 基础信息
             model_info = {
                 "id": model_id,
                 "object": "model", 
-                "created": current_timestamp,
+                "created": timestamp,
                 "owned_by": "x-ai",
                 "display_name": config.get("display_name", model_id),
                 "description": config.get("description", ""),
@@ -61,22 +41,16 @@ async def list_models(_: Optional[str] = Depends(auth_manager.verify)) -> Dict[s
             
             model_data.append(model_info)
         
-        # 构建响应
-        response = {
-            "object": "list",
-            "data": model_data
-        }
-
-        logger.debug(f"[Models] 成功返回 {len(model_data)} 个模型")
-        return response
+        logger.debug(f"[Models] 返回 {len(model_data)} 个模型")
+        return {"object": "list", "data": model_data}
         
     except Exception as e:
-        logger.error(f"[Models] 获取模型列表时发生错误: {str(e)}")
+        logger.error(f"[Models] 获取列表失败: {e}")
         raise HTTPException(
             status_code=500,
             detail={
                 "error": {
-                    "message": f"Failed to retrieve models: {str(e)}",
+                    "message": f"Failed to retrieve models: {e}",
                     "type": "internal_error",
                     "code": "model_list_error"
                 }
@@ -86,22 +60,13 @@ async def list_models(_: Optional[str] = Depends(auth_manager.verify)) -> Dict[s
 
 @router.get("/models/{model_id}")
 async def get_model(model_id: str, _: Optional[str] = Depends(auth_manager.verify)) -> Dict[str, Any]:
-    """
-    获取特定模型信息
-
-    Args:
-        model_id (str): 模型ID
-        _: 认证依赖（自动验证）
-
-    Returns:
-        Dict[str, Any]: 模型详细信息
-    """
+    """获取特定模型信息"""
     try:
-        logger.debug(f"[Models] 请求获取模型信息: {model_id}")
+        logger.debug(f"[Models] 请求模型: {model_id}")
 
-        # 验证模型是否存在
+        # 验证模型
         if not Models.is_valid_model(model_id):
-            logger.warning(f"[Models] 请求的模型不存在: {model_id}")
+            logger.warning(f"[Models] 模型不存在: {model_id}")
             raise HTTPException(
                 status_code=404,
                 detail={
@@ -113,17 +78,13 @@ async def get_model(model_id: str, _: Optional[str] = Depends(auth_manager.verif
                 }
             )
         
-        # 获取当前时间戳
-        current_timestamp = int(time.time())
-        
-        # 获取模型配置
+        timestamp = int(time.time())
         config = Models.get_model_info(model_id)
         
-        # 构建模型信息
         model_info = {
             "id": model_id,
             "object": "model",
-            "created": current_timestamp,
+            "created": timestamp,
             "owned_by": "x-ai",
             "display_name": config.get("display_name", model_id),
             "description": config.get("description", ""),
@@ -134,18 +95,18 @@ async def get_model(model_id: str, _: Optional[str] = Depends(auth_manager.verif
             "default_top_p": config.get("default_top_p", 0.95)
         }
 
-        logger.debug(f"[Models] 成功返回模型信息: {model_id}")
+        logger.debug(f"[Models] 返回模型: {model_id}")
         return model_info
         
     except HTTPException:
         raise
     except Exception as e:
-        logger.error(f"[Models] 获取模型信息时发生错误: {str(e)}")
+        logger.error(f"[Models] 获取模型失败: {e}")
         raise HTTPException(
             status_code=500,
             detail={
                 "error": {
-                    "message": f"Failed to retrieve model: {str(e)}",
+                    "message": f"Failed to retrieve model: {e}",
                     "type": "internal_error",
                     "code": "model_retrieve_error"
                 }
diff --git a/app/core/auth.py b/app/core/auth.py
index fffaa9e..9a95f69 100644
--- a/app/core/auth.py
+++ b/app/core/auth.py
@@ -1,53 +1,58 @@
-"""认证模块"""
+"""认证模块 - API令牌验证"""
 
 from typing import Optional
 from fastapi import Depends, HTTPException
 from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
+
 from app.core.config import setting
 from app.core.logger import logger
 
 
+# Bearer安全方案
 security = HTTPBearer(auto_error=False)
 
 
+def _build_error(message: str, code: str = "invalid_token") -> dict:
+    """构建认证错误"""
+    return {
+        "error": {
+            "message": message,
+            "type": "authentication_error",
+            "code": code
+        }
+    }
+
+
 class AuthManager:
-    """认证管理器"""
+    """认证管理器 - 验证API令牌"""
 
     @staticmethod
     def verify(credentials: Optional[HTTPAuthorizationCredentials] = Depends(security)) -> Optional[str]:
-        """验证认证令牌"""
+        """验证令牌"""
         api_key = setting.grok_config.get("api_key")
 
+        # 未设置时跳过
         if not api_key:
-            logger.debug("[Auth] 未设置API_KEY，跳过验证。")
+            logger.debug("[Auth] 未设置API_KEY，跳过验证")
             return credentials.credentials if credentials else None
 
+        # 检查令牌
         if not credentials:
             raise HTTPException(
                 status_code=401,
-                detail={
-                    "error": {
-                        "message": "缺少认证令牌",
-                        "type": "authentication_error",
-                        "code": "missing_token"
-                    }
-                }
+                detail=_build_error("缺少认证令牌", "missing_token")
             )
 
+        # 验证令牌
         if credentials.credentials != api_key:
             raise HTTPException(
                 status_code=401,
-                detail={
-                    "error": {
-                        "message": f"令牌无效，长度: {len(credentials.credentials)}",
-                        "type": "authentication_error",
-                        "code": "invalid_token"
-                    }
-                }
+                detail=_build_error(f"令牌无效，长度: {len(credentials.credentials)}", "invalid_token")
             )
 
-        logger.debug("[Auth] 令牌认证成功。")
+        logger.debug("[Auth] 令牌认证成功")
         return credentials.credentials
 
 
+# 全局实例
 auth_manager = AuthManager()
\ No newline at end of file
diff --git a/app/core/config.py b/app/core/config.py
index 01226f7..dc68617 100644
--- a/app/core/config.py
+++ b/app/core/config.py
@@ -1,72 +1,193 @@
-"""配置管理器"""
+"""配置管理器 - 管理应用配置的读写"""
 
 import toml
 from pathlib import Path
-from typing import Dict, Any
+from typing import Dict, Any, Optional, Literal
+
+
+# 默认配置
+DEFAULT_GROK = {
+    "api_key": "",
+    "proxy_url": "",
+    "proxy_pool_url": "",
+    "proxy_pool_interval": 300,
+    "cache_proxy_url": "",
+    "cf_clearance": "",
+    "x_statsig_id": "",
+    "dynamic_statsig": True,
+    "filtered_tags": "xaiartifact,xai:tool_usage_card",
+    "show_thinking": True,
+    "temporary": False,
+    "max_upload_concurrency": 20,
+    "max_request_concurrency": 100,
+    "stream_first_response_timeout": 30,
+    "stream_chunk_timeout": 120,
+    "stream_total_timeout": 600,
+    "retry_status_codes": [401, 429],  # 可重试的HTTP状态码
+}
+
+DEFAULT_GLOBAL = {
+    "base_url": "http://localhost:8000",
+    "log_level": "INFO",
+    "image_mode": "url",
+    "admin_password": "admin",
+    "admin_username": "admin",
+    "image_cache_max_size_mb": 512,
+    "video_cache_max_size_mb": 1024,
+    "max_upload_concurrency": 20,  # 最大并发上传数
+    "max_request_concurrency": 50,  # 最大并发请求数
+    "batch_save_interval": 1.0,  # 批量保存间隔（秒）
+    "batch_save_threshold": 10  # 触发批量保存的变更数阈值
+}
 
 
 class ConfigManager:
     """配置管理器"""
 
     def __init__(self) -> None:
-        """初始化"""
-
-        # 加载环境变量
+        """初始化配置"""
         self.config_path: Path = Path(__file__).parents[2] / "data" / "setting.toml"
+        self._storage: Optional[Any] = None
+        self._ensure_exists()
         self.global_config: Dict[str, Any] = self.load("global")
         self.grok_config: Dict[str, Any] = self.load("grok")
-        self._storage = None
+    
+    def _ensure_exists(self) -> None:
+        """确保配置存在"""
+        if not self.config_path.exists():
+            self.config_path.parent.mkdir(parents=True, exist_ok=True)
+            self._create_default()
+    
+    def _create_default(self) -> None:
+        """创建默认配置"""
+        default = {"grok": DEFAULT_GROK.copy(), "global": DEFAULT_GLOBAL.copy()}
+        with open(self.config_path, "w", encoding="utf-8") as f:
+            toml.dump(default, f)
+    
+    def _normalize_proxy(self, proxy: str) -> str:
+        """标准化代理URL（socks5:// → socks5h://）"""
+        if proxy and proxy.startswith("socks5://"):
+            return proxy.replace("socks5://", "socks5h://", 1)
+        return proxy
+    
+    def _normalize_cf(self, cf: str) -> str:
+        """标准化CF Clearance（自动添加前缀）"""
+        if cf and not cf.startswith("cf_clearance="):
+            return f"cf_clearance={cf}"
+        return cf
 
-    def set_storage(self, storage) -> None:
+    def set_storage(self, storage: Any) -> None:
         """设置存储实例"""
         self._storage = storage
 
-    def load(self, section: str) -> Dict[str, Any]:
-        """配置加载器"""
+    def load(self, section: Literal["global", "grok"]) -> Dict[str, Any]:
+        """加载配置节"""
         try:
             with open(self.config_path, "r", encoding="utf-8") as f:
-                return toml.load(f)[section]
+                config = toml.load(f)[section]
+
+            # 标准化Grok配置
+            if section == "grok":
+                if "proxy_url" in config:
+                    config["proxy_url"] = self._normalize_proxy(config["proxy_url"])
+                if "cf_clearance" in config:
+                    config["cf_clearance"] = self._normalize_cf(config["cf_clearance"])
+
+            return config
         except Exception as e:
-            raise Exception(f"[Setting] 配置加载失败: {e}")
+            raise Exception(f"[Setting] 配置加载失败: {e}") from e
     
     async def reload(self) -> None:
-        """重新加载配置（用于从存储同步后）"""
+        """重新加载配置"""
         self.global_config = self.load("global")
         self.grok_config = self.load("grok")
-
-    async def save(self, global_config: Dict[str, Any] = None, grok_config: Dict[str, Any] = None) -> None:
-        """保存配置到存储"""
-        if not self._storage:
-            # 如果没有设置存储，使用传统文件保存方式（向后兼容）
-            import aiofiles
-            async with aiofiles.open(self.config_path, "r", encoding="utf-8") as f:
-                content = await f.read()
-                config = toml.loads(content)
-            
-            if global_config:
-                config["global"].update(global_config)
-            if grok_config:
-                config["grok"].update(grok_config)
-            
-            async with aiofiles.open(self.config_path, "w", encoding="utf-8") as f:
-                await f.write(toml.dumps(config))
-        else:
-            # 使用存储抽象层
-            config_data = await self._storage.load_config()
-            
-            if global_config:
-                config_data["global"].update(global_config)
-            if grok_config:
-                config_data["grok"].update(grok_config)
-            
-            await self._storage.save_config(config_data)
+    
+    async def _save_file(self, updates: Dict[str, Dict[str, Any]]) -> None:
+        """保存到文件"""
+        import aiofiles
+        
+        async with aiofiles.open(self.config_path, "r", encoding="utf-8") as f:
+            config = toml.loads(await f.read())
+        
+        for section, data in updates.items():
+            if section in config:
+                config[section].update(data)
+        
+        async with aiofiles.open(self.config_path, "w", encoding="utf-8") as f:
+            await f.write(toml.dumps(config))
+    
+    async def _save_storage(self, updates: Dict[str, Dict[str, Any]]) -> None:
+        """保存到存储"""
+        config = await self._storage.load_config()
+        
+        for section, data in updates.items():
+            if section in config:
+                config[section].update(data)
+        
+        await self._storage.save_config(config)
+    
+    def _prepare_grok(self, grok: Dict[str, Any]) -> Dict[str, Any]:
+        """准备Grok配置（移除前缀）"""
+        processed = grok.copy()
+        if "cf_clearance" in processed:
+            cf = processed["cf_clearance"]
+            if cf and cf.startswith("cf_clearance="):
+                processed["cf_clearance"] = cf.replace("cf_clearance=", "", 1)
+        return processed
+
+    async def save(self, global_config: Optional[Dict[str, Any]] = None, grok_config: Optional[Dict[str, Any]] = None) -> None:
+        """保存配置"""
+        updates = {}
+        
+        if global_config:
+            updates["global"] = global_config
+        if grok_config:
+            updates["grok"] = self._prepare_grok(grok_config)
+        
+        # 选择存储方式
+        if self._storage:
+            await self._save_storage(updates)
+        else:
+            await self._save_file(updates)
         
-        # 重新加载配置
         await self.reload()
+    
+    async def get_proxy_async(self, proxy_type: Literal["service", "cache"] = "service") -> str:
+        """异步获取代理URL（支持代理池）
+        
+        Args:
+            proxy_type: 代理类型
+                - service: 服务代理（client/upload）
+                - cache: 缓存代理（cache）
+        """
+        from app.core.proxy_pool import proxy_pool
+        
+        if proxy_type == "cache":
+            cache_proxy = self.grok_config.get("cache_proxy_url", "")
+            if cache_proxy:
+                return cache_proxy
+        
+        # 从代理池获取
+        return await proxy_pool.get_proxy() or ""
+    
+    def get_proxy(self, proxy_type: Literal["service", "cache"] = "service") -> str:
+        """获取代理URL（同步方法，用于向后兼容）
+        
+        Args:
+            proxy_type: 代理类型
+                - service: 服务代理（client/upload）
+                - cache: 缓存代理（cache）
+        """
+        from app.core.proxy_pool import proxy_pool
+        
+        if proxy_type == "cache":
+            cache_proxy = self.grok_config.get("cache_proxy_url", "")
+            if cache_proxy:
+                return cache_proxy
+        
+        # 返回当前代理（如果是代理池，返回最后一次获取的）
+        return proxy_pool.get_current_proxy() or self.grok_config.get("proxy_url", "")
 
-# 全局设置
+
+# 全局实例
 setting = ConfigManager()
-
-if __name__ == "__main__":
-    print(setting.global_config)
-    print(setting.grok_config)
\ No newline at end of file
diff --git a/app/core/exception.py b/app/core/exception.py
index a5c179b..0f13a1d 100644
--- a/app/core/exception.py
+++ b/app/core/exception.py
@@ -1,4 +1,4 @@
-"""异常处理器"""
+"""异常处理器 - OpenAI兼容的错误响应"""
 
 from fastapi import Request, status
 from fastapi.responses import JSONResponse
@@ -6,22 +6,60 @@ from fastapi.exceptions import RequestValidationError
 from starlette.exceptions import HTTPException as StarletteHTTPException
 
 
-class GrokApiException(Exception):
-    """Grok API 业务异常"""
+# HTTP错误映射
+HTTP_ERROR_MAP = {
+    400: ("invalid_request_error", "请求格式错误或缺少必填参数"),
+    401: ("invalid_request_error", "令牌认证失败"),
+    403: ("permission_error", "没有权限访问此资源"),
+    404: ("invalid_request_error", "请求的资源不存在"),
+    429: ("rate_limit_error", "请求频率超出限制，请稍后再试"),
+    500: ("api_error", "内部服务器错误"),
+    503: ("api_error", "服务暂时不可用"),
+}
 
-    def __init__(self, message: str, error_code: str = None, details: dict = None):
+# Grok错误码映射
+GROK_STATUS_MAP = {
+    "NO_AUTH_TOKEN": status.HTTP_401_UNAUTHORIZED,
+    "INVALID_TOKEN": status.HTTP_401_UNAUTHORIZED,
+    "HTTP_ERROR": status.HTTP_502_BAD_GATEWAY,
+    "NETWORK_ERROR": status.HTTP_503_SERVICE_UNAVAILABLE,
+    "JSON_ERROR": status.HTTP_502_BAD_GATEWAY,
+    "API_ERROR": status.HTTP_502_BAD_GATEWAY,
+    "STREAM_ERROR": status.HTTP_502_BAD_GATEWAY,
+    "NO_RESPONSE": status.HTTP_502_BAD_GATEWAY,
+    "TOKEN_SAVE_ERROR": status.HTTP_500_INTERNAL_SERVER_ERROR,
+    "NO_AVAILABLE_TOKEN": status.HTTP_503_SERVICE_UNAVAILABLE,
+}
+
+GROK_TYPE_MAP = {
+    "NO_AUTH_TOKEN": "authentication_error",
+    "INVALID_TOKEN": "authentication_error",
+    "HTTP_ERROR": "api_error",
+    "NETWORK_ERROR": "api_error",
+    "JSON_ERROR": "api_error",
+    "API_ERROR": "api_error",
+    "STREAM_ERROR": "api_error",
+    "NO_RESPONSE": "api_error",
+    "TOKEN_SAVE_ERROR": "api_error",
+    "NO_AVAILABLE_TOKEN": "api_error",
+}
+
+
+class GrokApiException(Exception):
+    """Grok API业务异常"""
+
+    def __init__(self, message: str, error_code: str = None, details: dict = None, context: dict = None):
         self.message = message
         self.error_code = error_code
         self.details = details or {}
+        self.context = context or {}
         super().__init__(self.message)
 
 
 def build_error_response(message: str, error_type: str, code: str = None, param: str = None) -> dict:
     """构建OpenAI兼容的错误响应"""
-    error = {
-        "message": message,
-        "type": error_type,
-    }
+    error = {"message": message, "type": error_type}
+    
     if code:
         error["code"] = code
     if param:
@@ -32,18 +70,8 @@ def build_error_response(message: str, error_type: str, code: str = None, param:
 
 async def http_exception_handler(_: Request, exc: StarletteHTTPException) -> JSONResponse:
     """处理HTTP异常"""
-    error_map = {
-        400: ("invalid_request_error", "请求格式错误或缺少必填参数。"),
-        401: ("invalid_request_error", "令牌认证失败。"),
-        403: ("permission_error", "没有权限访问此资源。"),
-        404: ("invalid_request_error", "请求的资源不存在。"),
-        429: ("rate_limit_error", "请求频率超出限制，请稍后再试。"),
-        500: ("api_error", "内部服务器错误。"),
-        503: ("api_error", "服务暂时不可用。"),
-    }
-
-    error_type, default_message = error_map.get(exc.status_code, ("api_error", str(exc.detail)))
-    message = str(exc.detail) if exc.detail else default_message
+    error_type, default_msg = HTTP_ERROR_MAP.get(exc.status_code, ("api_error", str(exc.detail)))
+    message = str(exc.detail) if exc.detail else default_msg
 
     return JSONResponse(
         status_code=exc.status_code,
@@ -55,7 +83,7 @@ async def validation_exception_handler(_: Request, exc: RequestValidationError)
     """处理验证错误"""
     errors = exc.errors()
     param = errors[0]["loc"][-1] if errors and errors[0].get("loc") else None
-    message = errors[0]["msg"] if errors and errors[0].get("msg") else "请求参数错误。"
+    message = errors[0]["msg"] if errors and errors[0].get("msg") else "请求参数错误"
 
     return JSONResponse(
         status_code=status.HTTP_400_BAD_REQUEST,
@@ -64,36 +92,9 @@ async def validation_exception_handler(_: Request, exc: RequestValidationError)
 
 
 async def grok_api_exception_handler(_: Request, exc: GrokApiException) -> JSONResponse:
-    """处理Grok API业务异常"""
-    # 根据错误码映射HTTP状态码
-    status_code_map = {
-        "NO_AUTH_TOKEN": status.HTTP_401_UNAUTHORIZED,
-        "INVALID_TOKEN": status.HTTP_401_UNAUTHORIZED,
-        "HTTP_ERROR": status.HTTP_502_BAD_GATEWAY,
-        "NETWORK_ERROR": status.HTTP_503_SERVICE_UNAVAILABLE,
-        "JSON_ERROR": status.HTTP_502_BAD_GATEWAY,
-        "API_ERROR": status.HTTP_502_BAD_GATEWAY,
-        "STREAM_ERROR": status.HTTP_502_BAD_GATEWAY,
-        "NO_RESPONSE": status.HTTP_502_BAD_GATEWAY,
-        "TOKEN_SAVE_ERROR": status.HTTP_500_INTERNAL_SERVER_ERROR,
-        "NO_AVAILABLE_TOKEN": status.HTTP_503_SERVICE_UNAVAILABLE,
-    }
-
-    http_status = status_code_map.get(exc.error_code, status.HTTP_500_INTERNAL_SERVER_ERROR)
-    error_type_map = {
-        "NO_AUTH_TOKEN": "authentication_error",
-        "INVALID_TOKEN": "authentication_error",
-        "HTTP_ERROR": "api_error",
-        "NETWORK_ERROR": "api_error",
-        "JSON_ERROR": "api_error",
-        "API_ERROR": "api_error",
-        "STREAM_ERROR": "api_error",
-        "NO_RESPONSE": "api_error",
-        "TOKEN_SAVE_ERROR": "api_error",
-        "NO_AVAILABLE_TOKEN": "api_error",
-    }
-
-    error_type = error_type_map.get(exc.error_code, "api_error")
+    """处理Grok API异常"""
+    http_status = GROK_STATUS_MAP.get(exc.error_code, status.HTTP_500_INTERNAL_SERVER_ERROR)
+    error_type = GROK_TYPE_MAP.get(exc.error_code, "api_error")
 
     return JSONResponse(
         status_code=http_status,
@@ -105,15 +106,12 @@ async def global_exception_handler(_: Request, exc: Exception) -> JSONResponse:
     """处理未捕获异常"""
     return JSONResponse(
         status_code=status.HTTP_500_INTERNAL_SERVER_ERROR,
-        content=build_error_response(
-            "服务器遇到意外错误，请重试。",
-            "api_error"
-        )
+        content=build_error_response("服务器遇到意外错误，请重试", "api_error")
     )
 
 
 def register_exception_handlers(app) -> None:
-    """注册OpenAI兼容的异常处理器"""
+    """注册异常处理器"""
     app.add_exception_handler(StarletteHTTPException, http_exception_handler)
     app.add_exception_handler(RequestValidationError, validation_exception_handler)
     app.add_exception_handler(GrokApiException, grok_api_exception_handler)
diff --git a/app/core/logger.py b/app/core/logger.py
index 5fded40..8129ff8 100644
--- a/app/core/logger.py
+++ b/app/core/logger.py
@@ -1,63 +1,109 @@
-"""全局日志模块"""
+"""全局日志模块 - 单例模式的日志管理器"""
 
 import sys
 import logging
 from pathlib import Path
 from logging.handlers import RotatingFileHandler
+
 from app.core.config import setting
 
 
-class LoggerManager:
-    """日志管理器"""
+# 过滤模式
+FILTER_PATTERNS = [
+    "chunk: b'",  # SSE原始字节
+    "Got event:",  # SSE事件
+    "Closing",  # SSE关闭
+]
 
+
+class MCPLogFilter(logging.Filter):
+    """MCP日志过滤器 - 过滤大量数据的DEBUG日志"""
+
+    def filter(self, record: logging.LogRecord) -> bool:
+        """过滤日志"""
+        # 过滤SSE的DEBUG日志
+        if record.name == "sse_starlette.sse" and record.levelno == logging.DEBUG:
+            msg = record.getMessage()
+            return not any(p in msg for p in FILTER_PATTERNS)
+
+        # 过滤MCP streamable_http的DEBUG日志
+        if "mcp.server.streamable_http" in record.name and record.levelno == logging.DEBUG:
+            return False
+
+        return True
+
+
+class LoggerManager:
+    """日志管理器（单例）"""
+
+    _instance = None
     _initialized = False
 
+    def __new__(cls):
+        if cls._instance is None:
+            cls._instance = super().__new__(cls)
+        return cls._instance
+
     def __init__(self):
-        """初始化日志"""
+        """初始化日志系统"""
         if LoggerManager._initialized:
             return
 
-        # 日志配置
+        # 配置
         log_dir = Path(__file__).parents[2] / "logs"
         log_dir.mkdir(exist_ok=True)
         log_level = setting.global_config.get("log_level", "INFO").upper()
         log_format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
         log_file = log_dir / "app.log"
 
-        # 配置根日志器
+        # 根日志器
         self.logger = logging.getLogger()
         self.logger.setLevel(log_level)
 
-        # 避免重复添加处理器
+        # 避免重复添加
         if self.logger.handlers:
             return
 
-        # 创建格式器
+        # 格式器和过滤器
         formatter = logging.Formatter(log_format)
+        mcp_filter = MCPLogFilter()
 
         # 控制台处理器
-        console_handler = logging.StreamHandler(sys.stdout)
-        console_handler.setLevel(log_level)
-        console_handler.setFormatter(formatter)
+        console = logging.StreamHandler(sys.stdout)
+        console.setLevel(log_level)
+        console.setFormatter(formatter)
+        console.addFilter(mcp_filter)
 
-        # 文件处理器
+        # 文件处理器（10MB，5个备份）
         file_handler = RotatingFileHandler(
-            log_file, maxBytes=10 * 1024 * 1024, backupCount=5, encoding="utf-8"
+            log_file, maxBytes=10*1024*1024, backupCount=5, encoding="utf-8"
         )
         file_handler.setLevel(log_level)
         file_handler.setFormatter(formatter)
+        file_handler.addFilter(mcp_filter)
 
-        # 添加处理器到根日志器
-        self.logger.addHandler(console_handler)
+        # 添加处理器
+        self.logger.addHandler(console)
         self.logger.addHandler(file_handler)
 
-        # 配置第三方库日志级别，避免过多调试信息
-        logging.getLogger("asyncio").setLevel(logging.WARNING)
-        logging.getLogger("uvicorn").setLevel(logging.INFO)
-        logging.getLogger("fastapi").setLevel(logging.INFO)
-        logging.getLogger("aiomysql").setLevel(logging.WARNING)
+        # 配置第三方库
+        self._configure_third_party()
 
         LoggerManager._initialized = True
+    
+    def _configure_third_party(self):
+        """配置第三方库日志级别"""
+        config = {
+            "asyncio": logging.WARNING,
+            "uvicorn": logging.INFO,
+            "fastapi": logging.INFO,
+            "aiomysql": logging.WARNING,
+            "mcp": logging.CRITICAL,
+            "fastmcp": logging.CRITICAL,
+        }
+        
+        for name, level in config.items():
+            logging.getLogger(name).setLevel(level)
 
     def debug(self, msg: str) -> None:
         """调试日志"""
@@ -80,5 +126,5 @@ class LoggerManager:
         self.logger.critical(msg)
 
 
-# 全局日志器实例
+# 全局实例
 logger = LoggerManager()
diff --git a/app/core/proxy_pool.py b/app/core/proxy_pool.py
new file mode 100644
index 0000000..eaea5f5
--- /dev/null
+++ b/app/core/proxy_pool.py
@@ -0,0 +1,143 @@
+"""代理池管理器 - 从URL动态获取代理IP"""
+
+import asyncio
+import aiohttp
+import time
+from typing import Optional, List
+from app.core.logger import logger
+
+
+class ProxyPool:
+    """代理池管理器"""
+    
+    def __init__(self):
+        self._pool_url: Optional[str] = None
+        self._static_proxy: Optional[str] = None
+        self._current_proxy: Optional[str] = None
+        self._last_fetch_time: float = 0
+        self._fetch_interval: int = 300  # 5分钟刷新一次
+        self._enabled: bool = False
+        self._lock = asyncio.Lock()
+    
+    def configure(self, proxy_url: str, proxy_pool_url: str = "", proxy_pool_interval: int = 300):
+        """配置代理池
+        
+        Args:
+            proxy_url: 静态代理URL（socks5h://xxx 或 http://xxx）
+            proxy_pool_url: 代理池API URL，返回单个代理地址
+            proxy_pool_interval: 代理池刷新间隔（秒）
+        """
+        self._static_proxy = proxy_url if proxy_url else None
+        self._pool_url = proxy_pool_url if proxy_pool_url else None
+        self._fetch_interval = proxy_pool_interval
+        self._enabled = bool(self._pool_url)
+        
+        if self._enabled:
+            logger.info(f"[ProxyPool] 代理池已启用: {self._pool_url}, 刷新间隔: {self._fetch_interval}s")
+        elif self._static_proxy:
+            logger.info(f"[ProxyPool] 使用静态代理: {self._static_proxy}")
+            self._current_proxy = self._static_proxy
+        else:
+            logger.info("[ProxyPool] 未配置代理")
+    
+    async def get_proxy(self) -> Optional[str]:
+        """获取代理地址
+        
+        Returns:
+            代理URL或None
+        """
+        # 如果未启用代理池，返回静态代理
+        if not self._enabled:
+            return self._static_proxy
+        
+        # 检查是否需要刷新
+        now = time.time()
+        if not self._current_proxy or (now - self._last_fetch_time) >= self._fetch_interval:
+            async with self._lock:
+                # 双重检查
+                if not self._current_proxy or (now - self._last_fetch_time) >= self._fetch_interval:
+                    await self._fetch_proxy()
+        
+        return self._current_proxy
+    
+    async def force_refresh(self) -> Optional[str]:
+        """强制刷新代理（用于403错误重试）
+        
+        Returns:
+            新的代理URL或None
+        """
+        if not self._enabled:
+            return self._static_proxy
+        
+        async with self._lock:
+            await self._fetch_proxy()
+        
+        return self._current_proxy
+    
+    async def _fetch_proxy(self):
+        """从代理池URL获取新的代理"""
+        try:
+            logger.debug(f"[ProxyPool] 正在从代理池获取新代理: {self._pool_url}")
+            
+            timeout = aiohttp.ClientTimeout(total=10)
+            async with aiohttp.ClientSession(timeout=timeout) as session:
+                async with session.get(self._pool_url) as response:
+                    if response.status == 200:
+                        proxy_text = await response.text()
+                        proxy = proxy_text.strip()
+                        
+                        # 验证代理格式
+                        if self._validate_proxy(proxy):
+                            self._current_proxy = proxy
+                            self._last_fetch_time = time.time()
+                            logger.info(f"[ProxyPool] 成功获取新代理: {proxy}")
+                        else:
+                            logger.error(f"[ProxyPool] 代理格式无效: {proxy}")
+                            # 降级到静态代理
+                            if not self._current_proxy:
+                                self._current_proxy = self._static_proxy
+                    else:
+                        logger.error(f"[ProxyPool] 获取代理失败: HTTP {response.status}")
+                        # 降级到静态代理
+                        if not self._current_proxy:
+                            self._current_proxy = self._static_proxy
+        
+        except asyncio.TimeoutError:
+            logger.error("[ProxyPool] 获取代理超时")
+            if not self._current_proxy:
+                self._current_proxy = self._static_proxy
+        
+        except Exception as e:
+            logger.error(f"[ProxyPool] 获取代理异常: {e}")
+            # 降级到静态代理
+            if not self._current_proxy:
+                self._current_proxy = self._static_proxy
+    
+    def _validate_proxy(self, proxy: str) -> bool:
+        """验证代理格式
+        
+        Args:
+            proxy: 代理URL
+        
+        Returns:
+            是否有效
+        """
+        if not proxy:
+            return False
+        
+        # 支持的协议
+        valid_protocols = ['http://', 'https://', 'socks5://', 'socks5h://']
+        
+        return any(proxy.startswith(proto) for proto in valid_protocols)
+    
+    def get_current_proxy(self) -> Optional[str]:
+        """获取当前使用的代理（同步方法）
+        
+        Returns:
+            当前代理URL或None
+        """
+        return self._current_proxy or self._static_proxy
+
+
+# 全局代理池实例
+proxy_pool = ProxyPool()
diff --git a/app/core/storage.py b/app/core/storage.py
index 942ee3e..6ef93ec 100644
--- a/app/core/storage.py
+++ b/app/core/storage.py
@@ -1,7 +1,7 @@
 """存储抽象层 - 支持文件、MySQL和Redis存储"""
 
 import os
-import json
+import orjson
 import toml
 import asyncio
 import warnings
@@ -47,7 +47,7 @@ class BaseStorage(ABC):
 
 
 class FileStorage(BaseStorage):
-    """文件存储实现"""
+    """文件存储"""
 
     def __init__(self, data_dir: Path):
         self.data_dir = data_dir
@@ -60,89 +60,87 @@ class FileStorage(BaseStorage):
         """初始化文件存储"""
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
-        # 初始化token文件
         if not self.token_file.exists():
-            await self._write_file(self.token_file, json.dumps({"sso": {}, "ssoSuper": {}}, indent=2, ensure_ascii=False))
-            logger.info("[Storage] 创建新的token文件")
+            await self._write(self.token_file, orjson.dumps({"sso": {}, "ssoSuper": {}}, option=orjson.OPT_INDENT_2).decode())
+            logger.info("[Storage] 创建token文件")
 
-        # 初始化配置文件
         if not self.config_file.exists():
-            default_config = {
+            default = {
                 "global": {"api_keys": [], "admin_username": "admin", "admin_password": "admin"},
                 "grok": {"proxy_url": "", "cf_clearance": "", "x_statsig_id": ""}
             }
-            await self._write_file(self.config_file, toml.dumps(default_config))
-            logger.info("[Storage] 创建新的配置文件")
+            await self._write(self.config_file, toml.dumps(default))
+            logger.info("[Storage] 创建配置文件")
 
-    async def _read_file(self, file_path: Path) -> str:
-        """读取文件内容"""
-        async with aiofiles.open(file_path, "r", encoding="utf-8") as f:
+    async def _read(self, path: Path) -> str:
+        """读取文件"""
+        async with aiofiles.open(path, "r", encoding="utf-8") as f:
             return await f.read()
 
-    async def _write_file(self, file_path: Path, content: str) -> None:
-        """写入文件内容"""
-        async with aiofiles.open(file_path, "w", encoding="utf-8") as f:
+    async def _write(self, path: Path, content: str) -> None:
+        """写入文件"""
+        async with aiofiles.open(path, "w", encoding="utf-8") as f:
             await f.write(content)
 
-    async def _load_json(self, file_path: Path, default: Dict[str, Any], lock: asyncio.Lock) -> Dict[str, Any]:
-        """加载JSON文件"""
+    async def _load_json(self, path: Path, default: Dict, lock: asyncio.Lock) -> Dict[str, Any]:
+        """加载JSON"""
         try:
             async with lock:
-                if not file_path.exists():
+                if not path.exists():
                     return default
-                return json.loads(await self._read_file(file_path))
+                return orjson.loads(await self._read(path))
         except Exception as e:
-            logger.error(f"[Storage] 加载{file_path.name}失败: {e}")
+            logger.error(f"[Storage] 加载{path.name}失败: {e}")
             return default
 
-    async def _save_json(self, file_path: Path, data: Dict[str, Any], lock: asyncio.Lock) -> None:
-        """保存JSON文件"""
+    async def _save_json(self, path: Path, data: Dict, lock: asyncio.Lock) -> None:
+        """保存JSON"""
         try:
             async with lock:
-                await self._write_file(file_path, json.dumps(data, indent=2, ensure_ascii=False))
+                await self._write(path, orjson.dumps(data, option=orjson.OPT_INDENT_2).decode())
         except Exception as e:
-            logger.error(f"[Storage] 保存{file_path.name}失败: {e}")
+            logger.error(f"[Storage] 保存{path.name}失败: {e}")
             raise
 
-    async def _load_toml(self, file_path: Path, default: Dict[str, Any], lock: asyncio.Lock) -> Dict[str, Any]:
-        """加载TOML文件"""
+    async def _load_toml(self, path: Path, default: Dict, lock: asyncio.Lock) -> Dict[str, Any]:
+        """加载TOML"""
         try:
             async with lock:
-                if not file_path.exists():
+                if not path.exists():
                     return default
-                return toml.loads(await self._read_file(file_path))
+                return toml.loads(await self._read(path))
         except Exception as e:
-            logger.error(f"[Storage] 加载{file_path.name}失败: {e}")
+            logger.error(f"[Storage] 加载{path.name}失败: {e}")
             return default
 
-    async def _save_toml(self, file_path: Path, data: Dict[str, Any], lock: asyncio.Lock) -> None:
-        """保存TOML文件"""
+    async def _save_toml(self, path: Path, data: Dict, lock: asyncio.Lock) -> None:
+        """保存TOML"""
         try:
             async with lock:
-                await self._write_file(file_path, toml.dumps(data))
+                await self._write(path, toml.dumps(data))
         except Exception as e:
-            logger.error(f"[Storage] 保存{file_path.name}失败: {e}")
+            logger.error(f"[Storage] 保存{path.name}失败: {e}")
             raise
 
     async def load_tokens(self) -> Dict[str, Any]:
-        """加载token数据"""
+        """加载token"""
         return await self._load_json(self.token_file, {"sso": {}, "ssoSuper": {}}, self._token_lock)
 
     async def save_tokens(self, data: Dict[str, Any]) -> None:
-        """保存token数据"""
+        """保存token"""
         await self._save_json(self.token_file, data, self._token_lock)
 
     async def load_config(self) -> Dict[str, Any]:
-        """加载配置数据"""
+        """加载配置"""
         return await self._load_toml(self.config_file, {"global": {}, "grok": {}}, self._config_lock)
 
     async def save_config(self, data: Dict[str, Any]) -> None:
-        """保存配置数据"""
+        """保存配置"""
         await self._save_toml(self.config_file, data, self._config_lock)
 
 
 class MysqlStorage(BaseStorage):
-    """MySQL存储实现"""
+    """MySQL存储"""
 
     def __init__(self, database_url: str, data_dir: Path):
         self.database_url = database_url
@@ -155,22 +153,15 @@ class MysqlStorage(BaseStorage):
         try:
             import aiomysql
             parsed = self._parse_url(self.database_url)
-            logger.info(f"[Storage] 解析数据库连接: {parsed['user']}@{parsed['host']}:{parsed['port']}/{parsed['db']}")
+            logger.info(f"[Storage] MySQL: {parsed['user']}@{parsed['host']}:{parsed['port']}/{parsed['db']}")
 
-            # 创建数据库
             await self._create_db(parsed)
-
-            # 创建连接池
             self._pool = await aiomysql.create_pool(
                 host=parsed['host'], port=parsed['port'], user=parsed['user'],
                 password=parsed['password'], db=parsed['db'], charset="utf8mb4",
                 autocommit=True, maxsize=10
             )
-
-            # 创建表
             await self._create_tables()
-
-            # 初始化文件存储和同步数据
             await self._file.init_db()
             await self._sync_data()
 
@@ -181,26 +172,26 @@ class MysqlStorage(BaseStorage):
             raise
 
     def _parse_url(self, url: str) -> Dict[str, Any]:
-        """解析数据库URL"""
-        parsed = urlparse(url)
+        """解析URL"""
+        p = urlparse(url)
         return {
-            'user': unquote(parsed.username) if parsed.username else "",
-            'password': unquote(parsed.password) if parsed.password else "",
-            'host': parsed.hostname,
-            'port': parsed.port or 3306,
-            'db': parsed.path[1:] if parsed.path else "grok2api"
+            'user': unquote(p.username) if p.username else "",
+            'password': unquote(p.password) if p.password else "",
+            'host': p.hostname,
+            'port': p.port or 3306,
+            'db': p.path[1:] if p.path else "grok2api"
         }
 
-    async def _create_db(self, parsed: Dict[str, Any]) -> None:
+    async def _create_db(self, parsed: Dict) -> None:
         """创建数据库"""
         import aiomysql
-        temp_pool = await aiomysql.create_pool(
+        pool = await aiomysql.create_pool(
             host=parsed['host'], port=parsed['port'], user=parsed['user'],
             password=parsed['password'], charset="utf8mb4", autocommit=True, maxsize=1
         )
 
         try:
-            async with temp_pool.acquire() as conn:
+            async with pool.acquire() as conn:
                 async with conn.cursor() as cursor:
                     with warnings.catch_warnings():
                         warnings.filterwarnings('ignore', message='.*database exists')
@@ -208,10 +199,10 @@ class MysqlStorage(BaseStorage):
                             f"CREATE DATABASE IF NOT EXISTS `{parsed['db']}` "
                             f"CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci"
                         )
-                    logger.info(f"[Storage] MySQL数据库 '{parsed['db']}' 已就绪")
+                    logger.info(f"[Storage] 数据库 '{parsed['db']}' 就绪")
         finally:
-            temp_pool.close()
-            await temp_pool.wait_closed()
+            pool.close()
+            await pool.wait_closed()
 
     async def _create_tables(self) -> None:
         """创建表"""
@@ -240,7 +231,7 @@ class MysqlStorage(BaseStorage):
                     warnings.filterwarnings('ignore', message='.*already exists')
                     for sql in tables.values():
                         await cursor.execute(sql)
-                logger.info("[Storage] MySQL表创建/验证成功")
+                logger.info("[Storage] MySQL表就绪")
 
     async def _sync_data(self) -> None:
         """同步数据"""
@@ -252,39 +243,33 @@ class MysqlStorage(BaseStorage):
                         await self._file.save_tokens(data)
                     else:
                         await self._file.save_config(data)
-                    logger.info(f"[Storage] {table.split('_')[1]}数据已从数据库同步到文件")
+                    logger.info(f"[Storage] {table.split('_')[1]}数据已从DB同步")
                 else:
-                    if table == "grok_tokens":
-                        file_data = await self._file.load_tokens()
-                        if file_data.get(key) or file_data.get("ssoSuper"):
-                            await self._save_db(table, file_data)
-                            logger.info("[Storage] Token数据已从文件初始化到数据库")
-                    else:
-                        file_data = await self._file.load_config()
-                        if file_data.get(key) or file_data.get("grok"):
-                            await self._save_db(table, file_data)
-                            logger.info("[Storage] 配置数据已从文件初始化到数据库")
+                    file_data = await (self._file.load_tokens() if table == "grok_tokens" else self._file.load_config())
+                    if file_data.get(key) or (table == "grok_tokens" and file_data.get("ssoSuper")):
+                        await self._save_db(table, file_data)
+                        logger.info(f"[Storage] {table.split('_')[1]}数据已初始化到DB")
         except Exception as e:
-            logger.warning(f"[Storage] 数据同步失败: {e}")
+            logger.warning(f"[Storage] 同步失败: {e}")
 
-    async def _load_db(self, table: str) -> Optional[Dict[str, Any]]:
-        """从数据库加载数据"""
+    async def _load_db(self, table: str) -> Optional[Dict]:
+        """从DB加载"""
         try:
             async with self._pool.acquire() as conn:
                 async with conn.cursor() as cursor:
                     await cursor.execute(f"SELECT data FROM {table} ORDER BY id DESC LIMIT 1")
                     result = await cursor.fetchone()
-                    return json.loads(result[0]) if result else None
+                    return orjson.loads(result[0]) if result else None
         except Exception as e:
-            logger.error(f"[Storage] 从数据库加载{table}失败: {e}")
+            logger.error(f"[Storage] 加载{table}失败: {e}")
             return None
 
-    async def _save_db(self, table: str, data: Dict[str, Any]) -> None:
-        """保存数据到数据库"""
+    async def _save_db(self, table: str, data: Dict) -> None:
+        """保存到DB"""
         try:
             async with self._pool.acquire() as conn:
                 async with conn.cursor() as cursor:
-                    json_data = json.dumps(data, ensure_ascii=False)
+                    json_data = orjson.dumps(data).decode()
                     await cursor.execute(f"SELECT id FROM {table} ORDER BY id DESC LIMIT 1")
                     result = await cursor.fetchone()
 
@@ -293,24 +278,24 @@ class MysqlStorage(BaseStorage):
                     else:
                         await cursor.execute(f"INSERT INTO {table} (data) VALUES (%s)", (json_data,))
         except Exception as e:
-            logger.error(f"[Storage] 保存数据到{table}失败: {e}")
+            logger.error(f"[Storage] 保存{table}失败: {e}")
             raise
 
     async def load_tokens(self) -> Dict[str, Any]:
-        """加载token数据"""
+        """加载token"""
         return await self._file.load_tokens()
 
     async def save_tokens(self, data: Dict[str, Any]) -> None:
-        """保存token数据"""
+        """保存token"""
         await self._file.save_tokens(data)
         await self._save_db("grok_tokens", data)
 
     async def load_config(self) -> Dict[str, Any]:
-        """加载配置数据"""
+        """加载配置"""
         return await self._file.load_config()
 
     async def save_config(self, data: Dict[str, Any]) -> None:
-        """保存配置数据"""
+        """保存配置"""
         await self._file.save_config(data)
         await self._save_db("grok_settings", data)
 
@@ -319,11 +304,11 @@ class MysqlStorage(BaseStorage):
         if self._pool:
             self._pool.close()
             await self._pool.wait_closed()
-            logger.info("[Storage] MySQL连接池已关闭")
+            logger.info("[Storage] MySQL已关闭")
 
 
 class RedisStorage(BaseStorage):
-    """Redis存储实现"""
+    """Redis存储"""
 
     def __init__(self, redis_url: str, data_dir: Path):
         self.redis_url = redis_url
@@ -334,22 +319,18 @@ class RedisStorage(BaseStorage):
     async def init_db(self) -> None:
         """初始化Redis"""
         try:
-            import redis.asyncio as redis
-            parsed = self._parse_url(self.redis_url)
-            logger.info(f"[Storage] 解析Redis URL: host={parsed['host']}, port={parsed['port']}, db={parsed.get('db', 0)}, username={parsed.get('username')}, password={'***' if parsed.get('password') else None}")
+            import redis.asyncio as aioredis
+            parsed = urlparse(self.redis_url)
+            db = int(parsed.path.lstrip('/')) if parsed.path and parsed.path != '/' else 0
+            logger.info(f"[Storage] Redis: {parsed.hostname}:{parsed.port or 6379}/{db}")
 
-            # 创建Redis连接
-            self._redis = redis.Redis(
-                host=parsed['host'], port=parsed['port'], password=parsed.get('password'),
-                username=parsed.get('username'), db=parsed.get('db', 0),
-                encoding="utf-8", decode_responses=True
+            self._redis = aioredis.Redis.from_url(
+                self.redis_url, encoding="utf-8", decode_responses=True
             )
 
-            # 测试连接
             await self._redis.ping()
-            logger.info(f"[Storage] Redis连接成功: {parsed['host']}:{parsed['port']}/{parsed['db']}")
+            logger.info(f"[Storage] Redis连接成功")
 
-            # 初始化文件存储和同步数据
             await self._file.init_db()
             await self._sync_data()
 
@@ -359,26 +340,6 @@ class RedisStorage(BaseStorage):
             logger.error(f"[Storage] Redis初始化失败: {e}")
             raise
 
-    def _parse_url(self, url: str) -> Dict[str, Any]:
-        """解析Redis URL"""
-        if url.startswith('redis://'):
-            url = url[8:]
-        parsed = urlparse(f'//{url}')
-
-        result = {
-            'host': parsed.hostname or 'localhost',
-            'port': parsed.port or 6379,
-            'db': int(parsed.path.lstrip('/')) if parsed.path and parsed.path != '/' else 0,
-            'username': unquote(parsed.username) if parsed.username else None,
-            'password': unquote(parsed.password) if parsed.password else None
-        }
-
-        # Redis 6+ 默认用户名
-        if result['password'] and not result['username']:
-            result['username'] = 'default'
-
-        return result
-
     async def _sync_data(self) -> None:
         """同步数据"""
         try:
@@ -388,44 +349,43 @@ class RedisStorage(BaseStorage):
             ]:
                 data = await self._redis.get(key)
                 if data:
-                    parsed = json.loads(data)
+                    parsed = orjson.loads(data)
                     if key == "grok:tokens":
                         await self._file.save_tokens(parsed)
                     else:
                         await self._file.save_config(parsed)
-                    logger.info(f"[Storage] {key.split(':')[1]}数据已从Redis同步到文件")
+                    logger.info(f"[Storage] {key.split(':')[1]}数据已从Redis同步")
                 else:
                     file_data = await file_func()
                     if file_data.get(key_name) or (key == "grok:tokens" and file_data.get("ssoSuper")):
-                        json_data = json.dumps(file_data, ensure_ascii=False)
-                        await self._redis.set(key, json_data)
-                        logger.info(f"[Storage] {key.split(':')[1]}数据已从文件初始化到Redis")
+                        await self._redis.set(key, orjson.dumps(file_data).decode())
+                        logger.info(f"[Storage] {key.split(':')[1]}数据已初始化到Redis")
         except Exception as e:
-            logger.warning(f"[Storage] 数据同步失败: {e}")
+            logger.warning(f"[Storage] 同步失败: {e}")
 
-    async def _save_redis(self, key: str, data: Dict[str, Any]) -> None:
+    async def _save_redis(self, key: str, data: Dict) -> None:
         """保存到Redis"""
         try:
-            await self._redis.set(key, json.dumps(data, ensure_ascii=False))
+            await self._redis.set(key, orjson.dumps(data).decode())
         except Exception as e:
-            logger.error(f"[Storage] 保存到Redis失败: {e}")
+            logger.error(f"[Storage] 保存Redis失败: {e}")
             raise
 
     async def load_tokens(self) -> Dict[str, Any]:
-        """加载token数据"""
+        """加载token"""
         return await self._file.load_tokens()
 
     async def save_tokens(self, data: Dict[str, Any]) -> None:
-        """保存token数据"""
+        """保存token"""
         await self._file.save_tokens(data)
         await self._save_redis("grok:tokens", data)
 
     async def load_config(self) -> Dict[str, Any]:
-        """加载配置数据"""
+        """加载配置"""
         return await self._file.load_config()
 
     async def save_config(self, data: Dict[str, Any]) -> None:
-        """保存配置数据"""
+        """保存配置"""
         await self._file.save_config(data)
         await self._save_redis("grok:settings", data)
 
@@ -433,11 +393,11 @@ class RedisStorage(BaseStorage):
         """关闭连接"""
         if self._redis:
             await self._redis.close()
-            logger.info("[Storage] Redis连接已关闭")
+            logger.info("[Storage] Redis已关闭")
 
 
 class StorageManager:
-    """存储管理器"""
+    """存储管理器（单例）"""
 
     _instance: Optional['StorageManager'] = None
     _storage: Optional[BaseStorage] = None
@@ -457,22 +417,17 @@ class StorageManager:
         url = os.getenv("DATABASE_URL", "")
         data_dir = Path(__file__).parents[2] / "data"
 
-        storage_classes = {
-            "mysql": MysqlStorage,
-            "redis": RedisStorage,
-            "file": FileStorage
-        }
+        classes = {"mysql": MysqlStorage, "redis": RedisStorage, "file": FileStorage}
 
         if mode in ("mysql", "redis") and not url:
-            raise ValueError(f"{mode.upper()}模式需要DATABASE_URL环境变量")
+            raise ValueError(f"{mode.upper()}模式需要DATABASE_URL")
 
-        storage_class = storage_classes.get(mode, FileStorage)
+        storage_class = classes.get(mode, FileStorage)
         self._storage = storage_class(url, data_dir) if mode != "file" else storage_class(data_dir)
 
         await self._storage.init_db()
         self._initialized = True
-        logger.info(f"[Storage] 使用{mode}存储模式")
-        logger.info("[Storage] 存储管理器初始化完成")
+        logger.info(f"[Storage] 使用{mode}模式")
 
     def get_storage(self) -> BaseStorage:
         """获取存储实例"""
@@ -486,5 +441,5 @@ class StorageManager:
             await self._storage.close()
 
 
-# 全局存储管理器实例
+# 全局实例
 storage_manager = StorageManager()
diff --git a/app/models/grok_models.py b/app/models/grok_models.py
index 1f99064..b067c54 100644
--- a/app/models/grok_models.py
+++ b/app/models/grok_models.py
@@ -1,97 +1,130 @@
-from enum import Enum
+"""Grok 模型配置和枚举定义"""
 
-# 模型配置字典
-_MODEL_CONFIG = {
-        "grok-3-fast": {
-            "grok_model": ("grok-3", "MODEL_MODE_FAST"),
-            "rate_limit_model": "grok-3",
-            "cost": {"type": "low_cost", "multiplier": 1, "description": "计1次调用"},
-            "requires_super": False,
-            "display_name": "Grok 3 Fast",
-            "description": "Fast and efficient Grok 3 model",
-            "raw_model_path": "xai/grok-3",
-            "default_temperature": 1.0,
-            "default_max_output_tokens": 8192,
-            "supported_max_output_tokens": 131072,
-            "default_top_p": 0.95
-        },
-        "grok-4-fast": {
-            "grok_model": ("grok-4-mini-thinking-tahoe", "MODEL_MODE_GROK_4_MINI_THINKING"),
-            "rate_limit_model": "grok-4-mini-thinking-tahoe",
-            "cost": {"type": "low_cost", "multiplier": 1, "description": "计1次调用"},
-            "requires_super": False,
-            "display_name": "Grok 4 Fast",
-            "description": "Fast version of Grok 4 with mini thinking capabilities",
-            "raw_model_path": "xai/grok-4-mini-thinking-tahoe",
-            "default_temperature": 1.0,
-            "default_max_output_tokens": 8192,
-            "supported_max_output_tokens": 131072,
-            "default_top_p": 0.95
-        },
-        "grok-4-fast-expert": {
-            "grok_model": ("grok-4-mini-thinking-tahoe", "MODEL_MODE_EXPERT"),
-            "rate_limit_model": "grok-4-mini-thinking-tahoe",
-            "cost": {"type": "high_cost", "multiplier": 4, "description": "计4次调用"},
-            "requires_super": False,
-            "display_name": "Grok 4 Fast Expert",
-            "description": "Expert mode of Grok 4 Fast with enhanced reasoning",
-            "raw_model_path": "xai/grok-4-mini-thinking-tahoe",
-            "default_temperature": 1.0,
-            "default_max_output_tokens": 32768,
-            "supported_max_output_tokens": 131072,
-            "default_top_p": 0.95
-        },
-        "grok-4-expert": {
-            "grok_model": ("grok-4", "MODEL_MODE_EXPERT"),
-            "rate_limit_model": "grok-4",
-            "cost": {"type": "high_cost", "multiplier": 4, "description": "计4次调用"},
-            "requires_super": False,
-            "display_name": "Grok 4 Expert",
-            "description": "Full Grok 4 model with expert mode capabilities",
-            "raw_model_path": "xai/grok-4",
-            "default_temperature": 1.0,
-            "default_max_output_tokens": 32768,
-            "supported_max_output_tokens": 131072,
-            "default_top_p": 0.95
-        },
-        "grok-4-heavy": {
-            "grok_model": ("grok-4-heavy", "MODEL_MODE_HEAVY"),
-            "rate_limit_model": "grok-4-heavy",
-            "cost": {"type": "independent", "multiplier": 1, "description": "独立计费，只有Super用户可用"},
-            "requires_super": True,
-            "display_name": "Grok 4 Heavy",
-            "description": "Most powerful Grok 4 model with heavy computational capabilities. Requires Super Token for access.",
-            "raw_model_path": "xai/grok-4-heavy",
-            "default_temperature": 1.0,
-            "default_max_output_tokens": 65536,
-            "supported_max_output_tokens": 131072,
-            "default_top_p": 0.95
-        },
-        "grok-imagine-0.9": {
-            "grok_model": ("grok-3", "MODEL_MODE_FAST"),
-            "rate_limit_model": "grok-3",
-            "cost": {"type": "low_cost", "multiplier": 1, "description": "计1次调用"},
-            "requires_super": False,
-            "display_name": "Grok Imagine 0.9",
-            "description": "Video generation model powered by Grok",
-            "raw_model_path": "xai/grok-imagine-0.9",
-            "default_temperature": 1.0,
-            "default_max_output_tokens": 8192,
-            "supported_max_output_tokens": 131072,
-            "default_top_p": 0.95,
-            "is_video_model": True
-        }
+from enum import Enum
+from typing import Dict, Any, Tuple
+
+
+# 模型配置
+_MODEL_CONFIG: Dict[str, Dict[str, Any]] = {
+    "grok-3-fast": {
+        "grok_model": ("grok-3", "MODEL_MODE_FAST"),
+        "rate_limit_model": "grok-3",
+        "cost": {"type": "low_cost", "multiplier": 1, "description": "计1次调用"},
+        "requires_super": False,
+        "display_name": "Grok 3 Fast",
+        "description": "Fast and efficient Grok 3 model",
+        "raw_model_path": "xai/grok-3",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 8192,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95
+    },
+    "grok-4-fast": {
+        "grok_model": ("grok-4-mini-thinking-tahoe", "MODEL_MODE_GROK_4_MINI_THINKING"),
+        "rate_limit_model": "grok-4-mini-thinking-tahoe",
+        "cost": {"type": "low_cost", "multiplier": 1, "description": "计1次调用"},
+        "requires_super": False,
+        "display_name": "Grok 4 Fast",
+        "description": "Fast version of Grok 4 with mini thinking capabilities",
+        "raw_model_path": "xai/grok-4-mini-thinking-tahoe",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 8192,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95
+    },
+    "grok-4-fast-expert": {
+        "grok_model": ("grok-4-mini-thinking-tahoe", "MODEL_MODE_EXPERT"),
+        "rate_limit_model": "grok-4-mini-thinking-tahoe",
+        "cost": {"type": "high_cost", "multiplier": 4, "description": "计4次调用"},
+        "requires_super": False,
+        "display_name": "Grok 4 Fast Expert",
+        "description": "Expert mode of Grok 4 Fast with enhanced reasoning",
+        "raw_model_path": "xai/grok-4-mini-thinking-tahoe",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 32768,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95
+    },
+    "grok-4-expert": {
+        "grok_model": ("grok-4", "MODEL_MODE_EXPERT"),
+        "rate_limit_model": "grok-4",
+        "cost": {"type": "high_cost", "multiplier": 4, "description": "计4次调用"},
+        "requires_super": False,
+        "display_name": "Grok 4 Expert",
+        "description": "Full Grok 4 model with expert mode capabilities",
+        "raw_model_path": "xai/grok-4",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 32768,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95
+    },
+    "grok-4-heavy": {
+        "grok_model": ("grok-4-heavy", "MODEL_MODE_HEAVY"),
+        "rate_limit_model": "grok-4-heavy",
+        "cost": {"type": "independent", "multiplier": 1, "description": "独立计费，只有Super用户可用"},
+        "requires_super": True,
+        "display_name": "Grok 4 Heavy",
+        "description": "Most powerful Grok 4 model with heavy computational capabilities. Requires Super Token for access.",
+        "raw_model_path": "xai/grok-4-heavy",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 65536,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95
+    },
+    "grok-4.1": {
+        "grok_model": ("grok-4-1-non-thinking-w-tool", "MODEL_MODE_GROK_4_1"),
+        "rate_limit_model": "grok-4-1-non-thinking-w-tool",
+        "cost": {"type": "low_cost", "multiplier": 1, "description": "计1次调用"},
+        "requires_super": False,
+        "display_name": "Grok 4.1",
+        "description": "Latest Grok 4.1 model with tool capabilities",
+        "raw_model_path": "xai/grok-4-1-non-thinking-w-tool",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 8192,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95
+    },
+    "grok-4.1-thinking": {
+        "grok_model": ("grok-4-1-thinking-1108b", "MODEL_MODE_AUTO"),
+        "rate_limit_model": "grok-4-1-thinking-1108b",
+        "cost": {"type": "high_cost", "multiplier": 1, "description": "计1次调用"},
+        "requires_super": False,
+        "display_name": "Grok 4.1 Thinking",
+        "description": "Grok 4.1 model with advanced thinking and tool capabilities",
+        "raw_model_path": "xai/grok-4-1-thinking-1108b",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 32768,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95
+    },
+    "grok-imagine-0.9": {
+        "grok_model": ("grok-3", "MODEL_MODE_FAST"),
+        "rate_limit_model": "grok-3",
+        "cost": {"type": "low_cost", "multiplier": 1, "description": "计1次调用"},
+        "requires_super": False,
+        "display_name": "Grok Imagine 0.9",
+        "description": "Image and video generation model. Supports text-to-image and image-to-video generation.",
+        "raw_model_path": "xai/grok-imagine-0.9",
+        "default_temperature": 1.0,
+        "default_max_output_tokens": 8192,
+        "supported_max_output_tokens": 131072,
+        "default_top_p": 0.95,
+        "is_video_model": True
     }
+}
+
 
 class TokenType(Enum):
-    """Token类型枚举"""
-    NORMAL = "ssoNormal"  # 普通用户Token
-    SUPER = "ssoSuper"  # 超级用户Token
+    """Token类型"""
+    NORMAL = "ssoNormal"
+    SUPER = "ssoSuper"
 
 
 class Models(Enum):
-    """支持的模型枚举"""
+    """支持的模型"""
     GROK_3_FAST = "grok-3-fast"
+    GROK_4_1 = "grok-4.1"
+    GROK_4_1_THINKING = "grok-4.1-thinking"
     GROK_4_FAST = "grok-4-fast"
     GROK_4_FAST_EXPERT = "grok-4-fast-expert"
     GROK_4_EXPERT = "grok-4-expert"
@@ -99,8 +132,8 @@ class Models(Enum):
     GROK_IMAGINE_0_9 = "grok-imagine-0.9"
 
     @classmethod
-    def get_model_info(cls, model: str) -> dict:
-        """获取模型的完整配置信息"""
+    def get_model_info(cls, model: str) -> Dict[str, Any]:
+        """获取模型配置"""
         return _MODEL_CONFIG.get(model, {})
 
     @classmethod
@@ -109,15 +142,22 @@ class Models(Enum):
         return model in _MODEL_CONFIG
     
     @classmethod
-    def to_grok(cls, model: str) -> tuple[str, str]:
-        """转换为Grok内部模型名和模式类型"""
+    def to_grok(cls, model: str) -> Tuple[str, str]:
+        """转换为Grok内部模型名和模式
+        
+        Returns:
+            (模型名, 模式类型) 元组
+        """
         config = _MODEL_CONFIG.get(model)
-        if config:
-            return config["grok_model"]
-        return model, "MODEL_MODE_FAST"
+        return config["grok_model"] if config else (model, "MODEL_MODE_FAST")
     
     @classmethod
     def to_rate_limit(cls, model: str) -> str:
-        """转换为速率限制接口模型名"""
+        """转换为速率限制模型名"""
         config = _MODEL_CONFIG.get(model)
-        return config["rate_limit_model"] if config else model
\ No newline at end of file
+        return config["rate_limit_model"] if config else model
+    
+    @classmethod
+    def get_all_model_names(cls) -> list[str]:
+        """获取所有模型名称"""
+        return list(_MODEL_CONFIG.keys())
\ No newline at end of file
diff --git a/app/models/openai_schema.py b/app/models/openai_schema.py
index 37b6eeb..540eed9 100644
--- a/app/models/openai_schema.py
+++ b/app/models/openai_schema.py
@@ -1,16 +1,18 @@
-"""OpenAI 请求-响应模型"""
+"""OpenAI 请求-响应模型定义"""
 
 from fastapi import HTTPException
 from typing import Optional, List, Union, Dict, Any
 from pydantic import BaseModel, Field, field_validator
 
+from app.models.grok_models import Models
+
 
 class OpenAIChatRequest(BaseModel):
-    """OpenAI 聊天请求模型"""
+    """OpenAI聊天请求"""
 
     model: str = Field(..., description="模型名称", min_length=1)
     messages: List[Dict[str, Any]] = Field(..., description="消息列表", min_length=1)
-    stream: bool = Field(False, description="启用流式响应")
+    stream: bool = Field(False, description="流式响应")
     temperature: Optional[float] = Field(0.7, ge=0, le=2, description="采样温度")
     max_tokens: Optional[int] = Field(None, ge=1, le=100000, description="最大Token数")
     top_p: Optional[float] = Field(1.0, ge=0, le=1, description="采样参数")
@@ -20,31 +22,19 @@ class OpenAIChatRequest(BaseModel):
     def validate_messages(cls, v):
         """验证消息格式"""
         if not v:
-            raise HTTPException(
-                status_code=400,
-                detail="消息列表不能为空"
-            )
+            raise HTTPException(status_code=400, detail="消息列表不能为空")
 
         for msg in v:
             if not isinstance(msg, dict):
-                raise HTTPException(
-                    status_code=400,
-                    detail="每个消息必须是一个字典"
-                )
+                raise HTTPException(status_code=400, detail="每个消息必须是字典")
             if 'role' not in msg:
-                raise HTTPException(
-                    status_code=400,
-                    detail="消息缺少必填字段 'role'"
-                )
+                raise HTTPException(status_code=400, detail="消息缺少 'role' 字段")
             if 'content' not in msg:
-                raise HTTPException(
-                    status_code=400,
-                    detail="消息缺少必填字段 'content'"
-                )
+                raise HTTPException(status_code=400, detail="消息缺少 'content' 字段")
             if msg['role'] not in ['system', 'user', 'assistant']:
                 raise HTTPException(
                     status_code=400,
-                    detail=f"无效的角色 '{msg['role']}', 必须是 'system', 'user' 或 'assistant'"
+                    detail=f"无效角色 '{msg['role']}', 必须是 system/user/assistant"
                 )
 
         return v
@@ -53,29 +43,27 @@ class OpenAIChatRequest(BaseModel):
     @field_validator('model')
     def validate_model(cls, v):
         """验证模型名称"""
-        allowed_models = [
-            'grok-3-fast', 'grok-4-fast', 'grok-4-fast-expert',
-            'grok-4-expert', 'grok-4-heavy', 'grok-imagine-0.9'
-        ]
-        if v not in allowed_models:
+        if not Models.is_valid_model(v):
+            supported = Models.get_all_model_names()
             raise HTTPException(
                 status_code=400,
-                detail=f"不支持的模型 '{v}', 支持的模型: {', '.join(allowed_models)}"
+                detail=f"不支持的模型 '{v}', 支持: {', '.join(supported)}"
             )
         return v
 
+
 class OpenAIChatCompletionMessage(BaseModel):
     """聊天完成消息"""
     role: str = Field(..., description="角色")
-    content: str = Field(..., description="消息内容")
+    content: str = Field(..., description="内容")
     reference_id: Optional[str] = Field(default=None, description="参考ID")
     annotations: Optional[List[str]] = Field(default=None, description="注释")
 
 
 class OpenAIChatCompletionChoice(BaseModel):
     """聊天完成选项"""
-    index: int = Field(..., description="选项索引")
-    message: OpenAIChatCompletionMessage = Field(..., description="响应消息")
+    index: int = Field(..., description="索引")
+    message: OpenAIChatCompletionMessage = Field(..., description="消息")
     logprobs: Optional[float] = Field(default=None, description="对数概率")
     finish_reason: str = Field(default="stop", description="完成原因")
 
@@ -85,20 +73,20 @@ class OpenAIChatCompletionResponse(BaseModel):
     id: str = Field(..., description="响应ID")
     object: str = Field("chat.completion", description="对象类型")
     created: int = Field(..., description="创建时间戳")
-    model: str = Field(..., description="使用的模型")
-    choices: List[OpenAIChatCompletionChoice] = Field(..., description="响应选项")
+    model: str = Field(..., description="模型")
+    choices: List[OpenAIChatCompletionChoice] = Field(..., description="选项")
     usage: Optional[Dict[str, Any]] = Field(None, description="令牌使用")
 
 
 class OpenAIChatCompletionChunkMessage(BaseModel):
-    """流式响应消息片段"""
+    """流式消息片段"""
     role: str = Field(..., description="角色")
-    content: str = Field(..., description="消息内容")
+    content: str = Field(..., description="内容")
 
 
 class OpenAIChatCompletionChunkChoice(BaseModel):
-    """流式响应选项"""
-    index: int = Field(..., description="选项索引")
+    """流式选项"""
+    index: int = Field(..., description="索引")
     delta: Optional[Union[Dict[str, Any], OpenAIChatCompletionChunkMessage]] = Field(
         None, description="Delta数据"
     )
@@ -106,10 +94,10 @@ class OpenAIChatCompletionChunkChoice(BaseModel):
 
 
 class OpenAIChatCompletionChunkResponse(BaseModel):
-    """流式聊天完成响应"""
+    """流式聊天响应"""
     id: str = Field(..., description="响应ID")
     object: str = Field(default="chat.completion.chunk", description="对象类型")
     created: int = Field(..., description="创建时间戳")
-    model: str = Field(..., description="使用的模型")
+    model: str = Field(..., description="模型")
     system_fingerprint: Optional[str] = Field(default=None, description="系统指纹")
-    choices: List[OpenAIChatCompletionChunkChoice] = Field(..., description="响应选项")
\ No newline at end of file
+    choices: List[OpenAIChatCompletionChunkChoice] = Field(..., description="选项")
\ No newline at end of file
diff --git a/app/services/grok/cache.py b/app/services/grok/cache.py
index d9d9220..95e86b1 100644
--- a/app/services/grok/cache.py
+++ b/app/services/grok/cache.py
@@ -1,9 +1,9 @@
-"""缓存服务模块"""
+"""缓存服务模块 - 提供图片和视频的下载、缓存和清理功能"""
 
 import asyncio
 import base64
 from pathlib import Path
-from typing import Optional
+from typing import Optional, Tuple
 from curl_cffi.requests import AsyncSession
 
 from app.core.config import setting
@@ -11,166 +11,234 @@ from app.core.logger import logger
 from app.services.grok.statsig import get_dynamic_headers
 
 
+# 常量定义
+MIME_TYPES = {
+    '.jpg': 'image/jpeg', '.jpeg': 'image/jpeg', '.png': 'image/png',
+    '.gif': 'image/gif', '.webp': 'image/webp', '.bmp': 'image/bmp',
+}
+DEFAULT_MIME = 'image/jpeg'
+ASSETS_URL = "https://assets.grok.com"
+
+
 class CacheService:
     """缓存服务基类"""
 
-    def __init__(self, cache_type: str):
-        """初始化缓存服务"""
+    def __init__(self, cache_type: str, timeout: float = 30.0):
         self.cache_type = cache_type
         self.cache_dir = Path(f"data/temp/{cache_type}")
         self.cache_dir.mkdir(parents=True, exist_ok=True)
+        self.timeout = timeout
+        self._cleanup_lock = asyncio.Lock()
 
-    def _cache_path(self, file_path: str) -> Path:
-        """获取缓存文件的完整路径"""
-        filename = file_path.lstrip('/').replace('/', '-')
-        return self.cache_dir / filename
+    def _get_path(self, file_path: str) -> Path:
+        """转换文件路径为缓存路径"""
+        return self.cache_dir / file_path.lstrip('/').replace('/', '-')
 
-    async def download_file(self, file_path: str, auth_token: str, timeout: float = 30.0) -> Optional[Path]:
+    def _log(self, level: str, msg: str):
+        """统一日志输出"""
+        getattr(logger, level)(f"[{self.cache_type.upper()}Cache] {msg}")
+
+    def _build_headers(self, file_path: str, auth_token: str) -> dict:
+        """构建请求头"""
+        cf = setting.grok_config.get("cf_clearance", "")
+        return {
+            **get_dynamic_headers(pathname=file_path),
+            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8",
+            "Sec-Fetch-Dest": "document",
+            "Sec-Fetch-Mode": "navigate",
+            "Sec-Fetch-Site": "same-site",
+            "Sec-Fetch-User": "?1",
+            "Upgrade-Insecure-Requests": "1",
+            "Referer": "https://grok.com/",
+            "Cookie": f"{auth_token};{cf}" if cf else auth_token
+        }
+
+    async def download(self, file_path: str, auth_token: str, timeout: Optional[float] = None) -> Optional[Path]:
         """下载并缓存文件"""
-        cache_path = self._cache_path(file_path)
+        cache_path = self._get_path(file_path)
         if cache_path.exists():
-            logger.debug(f"[{self.cache_type.upper()}Cache] 文件已缓存: {cache_path}")
+            self._log("debug", "文件已缓存")
             return cache_path
 
-        try:
-            cf_clearance = setting.grok_config.get("cf_clearance", "")
-            headers = {
-                **get_dynamic_headers(pathname=file_path),
-                "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7",
-                "Sec-Fetch-Dest": "document",
-                "Sec-Fetch-Mode": "navigate",
-                "Sec-Fetch-Site": "same-site",
-                "Sec-Fetch-User": "?1",
-                "Upgrade-Insecure-Requests": "1",
-                "Referer": "https://grok.com/",
-                "Cookie": f"{auth_token};{cf_clearance}" if cf_clearance else auth_token
-            }
+        # 外层重试：可配置状态码（401/429等）
+        retry_codes = setting.grok_config.get("retry_status_codes", [401, 429])
+        MAX_OUTER_RETRY = 3
+        
+        for outer_retry in range(MAX_OUTER_RETRY + 1):  # +1 确保实际重试3次
+            try:
+                # 内层重试：403代理池重试（cache使用缓存代理，不支持代理池，所以403只重试一次）
+                max_403_retries = 5
+                retry_403_count = 0
+                
+                while retry_403_count <= max_403_retries:
+                    proxy = setting.get_proxy("cache")
+                    proxies = {"http": proxy, "https": proxy} if proxy else {}
+                    
+                    if proxy and outer_retry == 0 and retry_403_count == 0:
+                        self._log("debug", f"使用代理: {proxy.split('@')[-1] if '@' in proxy else proxy}")
 
-            proxy_url = setting.grok_config.get("proxy_url")
-            proxies = {"http": proxy_url, "https": proxy_url} if proxy_url else {}
-
-            async with AsyncSession() as session:
-                logger.debug(f"[{self.cache_type.upper()}Cache] 开始下载: https://assets.grok.com{file_path}")
-                response = await session.get(
-                    f"https://assets.grok.com{file_path}",
-                    headers=headers,
-                    proxies=proxies,
-                    timeout=timeout,
-                    allow_redirects=True,
-                    impersonate="chrome133a"
-                )
-                response.raise_for_status()
-                cache_path.write_bytes(response.content)
-                logger.debug(f"[{self.cache_type.upper()}Cache] 文件已缓存: {cache_path} ({len(response.content)} bytes)")
-                asyncio.create_task(self.cleanup_cache())
-                return cache_path
-        except Exception as e:
-            logger.error(f"[{self.cache_type.upper()}Cache] 下载文件失败: {e}")
-            return None
+                    async with AsyncSession() as session:
+                        url = f"{ASSETS_URL}{file_path}"
+                        if outer_retry == 0 and retry_403_count == 0:
+                            self._log("debug", f"下载: {url}")
+                        
+                        response = await session.get(
+                            url,
+                            headers=self._build_headers(file_path, auth_token),
+                            proxies=proxies,
+                            timeout=timeout or self.timeout,
+                            allow_redirects=True,
+                            impersonate="chrome133a"
+                        )
+                        
+                        # 检查403错误 - 内层重试(cache不使用代理池，所以直接失败)
+                        if response.status_code == 403:
+                            retry_403_count += 1
+                            
+                            if retry_403_count <= max_403_retries:
+                                self._log("warning", f"遇到403错误，正在重试 ({retry_403_count}/{max_403_retries})...")
+                                await asyncio.sleep(0.5)
+                                continue
+                            
+                            self._log("error", f"403错误，已重试{retry_403_count-1}次，放弃")
+                            return None
+                        
+                        # 检查可配置状态码错误 - 外层重试
+                        if response.status_code in retry_codes:
+                            if outer_retry < MAX_OUTER_RETRY:
+                                delay = (outer_retry + 1) * 0.1  # 渐进延迟：0.1s, 0.2s, 0.3s
+                                self._log("warning", f"遇到{response.status_code}错误，外层重试 ({outer_retry+1}/{MAX_OUTER_RETRY})，等待{delay}s...")
+                                await asyncio.sleep(delay)
+                                break  # 跳出内层循环，进入外层重试
+                            else:
+                                self._log("error", f"{response.status_code}错误，已重试{outer_retry}次，放弃")
+                                return None
+                        
+                        response.raise_for_status()
+                        
+                        cache_path.write_bytes(response.content)
+                        
+                        if outer_retry > 0 or retry_403_count > 0:
+                            self._log("info", f"重试成功！")
+                        else:
+                            self._log("debug", "缓存成功")
+                        
+                        # 异步清理（带错误处理）
+                        asyncio.create_task(self._safe_cleanup())
+                        return cache_path
+                        
+            except Exception as e:
+                if outer_retry < MAX_OUTER_RETRY - 1:
+                    self._log("warning", f"下载异常: {e}，外层重试 ({outer_retry+1}/{MAX_OUTER_RETRY})...")
+                    await asyncio.sleep(0.5)
+                    continue
+                
+                self._log("error", f"下载失败: {e}（已重试{outer_retry}次）")
+                return None
+        
+        return None
 
     def get_cached(self, file_path: str) -> Optional[Path]:
-        """获取缓存的文件路径"""
-        cache_path = self._cache_path(file_path)
-        return cache_path if cache_path.exists() else None
+        """获取已缓存的文件"""
+        path = self._get_path(file_path)
+        return path if path.exists() else None
 
-    async def cleanup_cache(self):
-        """清理缓存目录，确保不超过配置的大小限制"""
+    async def _safe_cleanup(self):
+        """安全清理（捕获异常）"""
         try:
-            max_size_mb = setting.global_config.get(f"{self.cache_type}_cache_max_size_mb", 500)
-            max_size_bytes = max_size_mb * 1024 * 1024
-
-            files = [(fp, (stat := fp.stat()).st_size, stat.st_mtime)
-                     for fp in self.cache_dir.glob("*") if fp.is_file()]
-            total_size = sum(size for _, size, _ in files)
-
-            if total_size <= max_size_bytes:
-                logger.debug(f"[{self.cache_type.upper()}Cache] 缓存大小 {total_size / 1024 / 1024:.2f}MB，未超限")
-                return
-
-            logger.info(f"[{self.cache_type.upper()}Cache] 缓存大小 {total_size / 1024 / 1024:.2f}MB 超过限制 {max_size_mb}MB，开始清理")
-            files.sort(key=lambda x: x[2])
-
-            for file_path, size, _ in files:
-                if total_size <= max_size_bytes:
-                    break
-                file_path.unlink()
-                total_size -= size
-                logger.debug(f"[{self.cache_type.upper()}Cache] 已删除缓存文件: {file_path}")
-
-            logger.info(f"[{self.cache_type.upper()}Cache] 缓存清理完成，当前大小 {total_size / 1024 / 1024:.2f}MB")
+            await self.cleanup()
         except Exception as e:
-            logger.error(f"[{self.cache_type.upper()}Cache] 清理缓存失败: {e}")
+            self._log("error", f"后台清理失败: {e}")
+
+    async def cleanup(self):
+        """清理超限缓存"""
+        if self._cleanup_lock.locked():
+            return
+        
+        async with self._cleanup_lock:
+            try:
+                max_mb = setting.global_config.get(f"{self.cache_type}_cache_max_size_mb", 500)
+                max_bytes = max_mb * 1024 * 1024
+
+                # 获取文件信息 (path, size, mtime)
+                files = [(f, (s := f.stat()).st_size, s.st_mtime) 
+                        for f in self.cache_dir.glob("*") if f.is_file()]
+                total = sum(size for _, size, _ in files)
+
+                if total <= max_bytes:
+                    return
+
+                self._log("info", f"清理缓存 {total/1024/1024:.1f}MB -> {max_mb}MB")
+                
+                # 删除最旧的文件
+                for path, size, _ in sorted(files, key=lambda x: x[2]):
+                    if total <= max_bytes:
+                        break
+                    path.unlink()
+                    total -= size
+                
+                self._log("info", f"清理完成: {total/1024/1024:.1f}MB")
+            except Exception as e:
+                self._log("error", f"清理失败: {e}")
 
 
-class ImageCacheService(CacheService):
+class ImageCache(CacheService):
     """图片缓存服务"""
 
     def __init__(self):
-        super().__init__("image")
+        super().__init__("image", timeout=30.0)
 
-    async def download_image(self, image_path: str, auth_token: str) -> Optional[Path]:
-        """下载并缓存图片"""
-        return await self.download_file(image_path, auth_token, timeout=30.0)
-
-    def get_cached(self, image_path: str) -> Optional[Path]:
-        """获取缓存的图片路径"""
-        return super().get_cached(image_path)
+    async def download_image(self, path: str, token: str) -> Optional[Path]:
+        """下载图片"""
+        return await self.download(path, token)
 
     @staticmethod
     def to_base64(image_path: Path) -> Optional[str]:
-        """将图片转换为 base64 编码"""
+        """图片转base64"""
         try:
             if not image_path.exists():
-                logger.error(f"[ImageCache] 图片文件不存在: {image_path}")
+                logger.error(f"[ImageCache] 文件不存在: {image_path}")
                 return None
 
-            with open(image_path, "rb") as f:
-                base64_data = base64.b64encode(f.read()).decode('utf-8')
-
-            mime_type = {'.jpg': 'image/jpeg', '.jpeg': 'image/jpeg', '.png': 'image/png',
-                         '.gif': 'image/gif', '.webp': 'image/webp'}.get(image_path.suffix.lower(), 'image/jpeg')
-
-            return f"data:{mime_type};base64,{base64_data}"
+            data = base64.b64encode(image_path.read_bytes()).decode()
+            mime = MIME_TYPES.get(image_path.suffix.lower(), DEFAULT_MIME)
+            return f"data:{mime};base64,{data}"
         except Exception as e:
-            logger.error(f"[ImageCache] 图片转 base64 失败: {e}")
+            logger.error(f"[ImageCache] 转换失败: {e}")
             return None
 
-    async def download_base64(self, image_path: str, auth_token: str) -> Optional[str]:
-        """下载图片并转换为 base64 编码（转换后立即删除缓存文件）"""
+    async def download_base64(self, path: str, token: str) -> Optional[str]:
+        """下载并转为base64（自动删除临时文件）"""
         try:
-            cache_path = await self.download_file(image_path, auth_token, timeout=30.0)
+            cache_path = await self.download(path, token)
             if not cache_path:
                 return None
 
-            base64_str = self.to_base64(cache_path)
-
+            result = self.to_base64(cache_path)
+            
+            # 清理临时文件
             try:
                 cache_path.unlink()
-                logger.debug(f"[ImageCache] 已删除临时文件: {cache_path}")
             except Exception as e:
                 logger.warning(f"[ImageCache] 删除临时文件失败: {e}")
 
-            return base64_str
+            return result
         except Exception as e:
-            logger.error(f"[ImageCache] 下载并转换 base64 失败: {e}")
+            logger.error(f"[ImageCache] 下载base64失败: {e}")
             return None
 
 
-class VideoCacheService(CacheService):
+class VideoCache(CacheService):
     """视频缓存服务"""
 
     def __init__(self):
-        super().__init__("video")
+        super().__init__("video", timeout=60.0)
 
-    async def download_video(self, video_path: str, auth_token: str) -> Optional[Path]:
-        """下载并缓存视频"""
-        return await self.download_file(video_path, auth_token, timeout=60.0)
-
-    def get_cached(self, video_path: str) -> Optional[Path]:
-        """获取缓存的视频路径"""
-        return super().get_cached(video_path)
+    async def download_video(self, path: str, token: str) -> Optional[Path]:
+        """下载视频"""
+        return await self.download(path, token)
 
 
 # 全局实例
-image_cache_service = ImageCacheService()
-video_cache_service = VideoCacheService()
+image_cache_service = ImageCache()
+video_cache_service = VideoCache()
diff --git a/app/services/grok/client.py b/app/services/grok/client.py
index b275308..0403de7 100644
--- a/app/services/grok/client.py
+++ b/app/services/grok/client.py
@@ -1,9 +1,8 @@
-"""Grok API 客户端模块"""
+"""Grok API 客户端 - 处理OpenAI到Grok的请求转换和响应处理"""
 
 import asyncio
-import json
-from typing import Dict, List, Tuple, Any
-
+import orjson
+from typing import Dict, List, Tuple, Any, Optional
 from curl_cffi import requests as curl_requests
 
 from app.core.config import setting
@@ -13,103 +12,163 @@ from app.services.grok.processer import GrokResponseProcessor
 from app.services.grok.statsig import get_dynamic_headers
 from app.services.grok.token import token_manager
 from app.services.grok.upload import ImageUploadManager
+from app.services.grok.create import PostCreateManager
 from app.core.exception import GrokApiException
 
-# 常量定义
-GROK_API_ENDPOINT = "https://grok.com/rest/app-chat/conversations/new"
-REQUEST_TIMEOUT = 120
-IMPERSONATE_BROWSER = "chrome133a"
+
+# 常量
+API_ENDPOINT = "https://grok.com/rest/app-chat/conversations/new"
+TIMEOUT = 120
+BROWSER = "chrome133a"
+MAX_RETRY = 3
+MAX_UPLOADS = 20  # 提高并发上传限制以支持更高并发
 
 
 class GrokClient:
     """Grok API 客户端"""
+    
+    _upload_sem = None  # 延迟初始化
 
     @staticmethod
-    async def openai_to_grok(openai_request: dict):
-        """转换OpenAI请求为Grok请求并处理响应"""
-        model = openai_request["model"]
-        messages = openai_request["messages"]
-        stream = openai_request.get("stream", False)
+    def _get_upload_semaphore():
+        """获取上传信号量（动态配置）"""
+        if GrokClient._upload_sem is None:
+            # 从配置读取，如果不可用则使用默认值
+            max_concurrency = setting.global_config.get("max_upload_concurrency", MAX_UPLOADS)
+            GrokClient._upload_sem = asyncio.Semaphore(max_concurrency)
+            logger.debug(f"[Client] 初始化上传并发限制: {max_concurrency}")
+        return GrokClient._upload_sem
 
-        logger.debug(f"[Client] 处理请求 - 模型:{model}, 消息数:{len(messages)}, 流式:{stream}")
-
-        # 提取消息内容和图片URL
-        content, image_urls = GrokClient._extract_content(messages)
-
-        # 获取认证令牌和模型信息
-        auth_token = token_manager.get_token(model)
-        model_name, model_mode = Models.to_grok(model)
-
-        # 检查是否为视频模型
-        is_video_model = Models.get_model_info(model).get("is_video_model", False)
+    @staticmethod
+    async def openai_to_grok(request: dict):
+        """转换OpenAI请求为Grok请求"""
+        model = request["model"]
+        content, images = GrokClient._extract_content(request["messages"])
+        stream = request.get("stream", False)
         
-        # 视频模型特殊处理：只允许一张图片
-        if is_video_model and len(image_urls) > 1:
-            logger.warning(f"[Client] 视频模型只允许一张图片，当前有{len(image_urls)}张，只使用第一张")
-            image_urls = image_urls[:1]
+        # 获取模型信息
+        info = Models.get_model_info(model)
+        grok_model, mode = Models.to_grok(model)
+        is_video = info.get("is_video_model", False)
         
-        # 上传图片并获取附件ID列表
-        image_attachments = await GrokClient._upload_imgs(image_urls, auth_token)
+        # 视频模型限制
+        if is_video and len(images) > 1:
+            logger.warning(f"[Client] 视频模型仅支持1张图片，已截取前1张")
+            images = images[:1]
+        
+        return await GrokClient._retry(model, content, images, grok_model, mode, is_video, stream)
 
-        # 视频模型：文本添加 --mode=custom
-        if is_video_model:
-            content = f"{content} --mode=custom"
-            logger.debug(f"[Client] 视频模型文本处理: {content}")
+    @staticmethod
+    async def _retry(model: str, content: str, images: List[str], grok_model: str, mode: str, is_video: bool, stream: bool):
+        """重试请求"""
+        last_err = None
 
-        # 构建Grok请求载荷
-        payload = GrokClient._build_payload(content, model_name, model_mode, image_attachments, is_video_model)
+        for i in range(MAX_RETRY):
+            try:
+                token = token_manager.get_token(model)
+                img_ids, img_uris = await GrokClient._upload(images, token)
 
-        return await GrokClient._send_request(payload, auth_token, model, stream)
+                # 视频模型创建会话
+                post_id = None
+                if is_video and img_ids and img_uris:
+                    post_id = await GrokClient._create_post(img_ids[0], img_uris[0], token)
+
+                payload = GrokClient._build_payload(content, grok_model, mode, img_ids, img_uris, is_video, post_id)
+                return await GrokClient._request(payload, token, model, stream, post_id)
+
+            except GrokApiException as e:
+                last_err = e
+                # 检查是否可重试
+                if e.error_code not in ["HTTP_ERROR", "NO_AVAILABLE_TOKEN"]:
+                    raise
+
+                status = e.context.get("status") if e.context else None
+                retry_codes = setting.grok_config.get("retry_status_codes", [401, 429])
+                
+                if status not in retry_codes:
+                    raise
+
+                if i < MAX_RETRY - 1:
+                    logger.warning(f"[Client] 失败(状态:{status}), 重试 {i+1}/{MAX_RETRY}")
+                    await asyncio.sleep(0.5)
+
+        raise last_err or GrokApiException("请求失败", "REQUEST_ERROR")
 
     @staticmethod
     def _extract_content(messages: List[Dict]) -> Tuple[str, List[str]]:
-        """提取消息内容和图片URL"""
-        content_parts = []
-        image_urls = []
-
+        """提取文本和图片"""
+        texts, images = [], []
+        
         for msg in messages:
-            msg_content = msg.get("content", "")
-
-            # 处理复杂消息格式（包含文本和图片）
-            if isinstance(msg_content, list):
-                for item in msg_content:
-                    item_type = item.get("type")
-                    if item_type == "text":
-                        content_parts.append(item.get("text", ""))
-                    elif item_type == "image_url":
-                        url = item.get("image_url", {}).get("url", "")
-                        if url:
-                            image_urls.append(url)
-            # 处理纯文本消息
+            content = msg.get("content", "")
+            
+            if isinstance(content, list):
+                for item in content:
+                    if item.get("type") == "text":
+                        texts.append(item.get("text", ""))
+                    elif item.get("type") == "image_url":
+                        if url := item.get("image_url", {}).get("url"):
+                            images.append(url)
             else:
-                content_parts.append(msg_content)
-
-        return "".join(content_parts), image_urls
+                texts.append(content)
+        
+        return "".join(texts), images
 
     @staticmethod
-    async def _upload_imgs(image_urls: List[str], auth_token: str) -> List[str]:
-        """上传图片并返回附件ID列表"""
-        image_attachments = []
-        # 并发上传所有图片
-        tasks = [ImageUploadManager.upload(url, auth_token) for url in image_urls]
-        results = await asyncio.gather(*tasks, return_exceptions=True)
-
-        for url, result in zip(image_urls, results):
+    async def _upload(urls: List[str], token: str) -> Tuple[List[str], List[str]]:
+        """并发上传图片"""
+        if not urls:
+            return [], []
+        
+        async def upload_limited(url):
+            async with GrokClient._get_upload_semaphore():
+                return await ImageUploadManager.upload(url, token)
+        
+        results = await asyncio.gather(*[upload_limited(u) for u in urls], return_exceptions=True)
+        
+        ids, uris = [], []
+        for url, result in zip(urls, results):
             if isinstance(result, Exception):
-                logger.warning(f"[Client] 图片上传失败: {url}, 错误: {result}")
-            elif result:
-                image_attachments.append(result)
-
-        return image_attachments
+                logger.warning(f"[Client] 上传失败: {url} - {result}")
+            elif isinstance(result, tuple) and len(result) == 2:
+                fid, furi = result
+                if fid:
+                    ids.append(fid)
+                    uris.append(furi)
+        
+        return ids, uris
 
     @staticmethod
-    def _build_payload(content: str, model_name: str, model_mode: str, image_attachments: List[str], is_video_model: bool = False) -> Dict[str, Any]:
-        """构建Grok API请求载荷"""
-        payload = {
+    async def _create_post(file_id: str, file_uri: str, token: str) -> Optional[str]:
+        """创建视频会话"""
+        try:
+            result = await PostCreateManager.create(file_id, file_uri, token)
+            if result and result.get("success"):
+                return result.get("post_id")
+        except Exception as e:
+            logger.warning(f"[Client] 创建会话失败: {e}")
+        return None
+
+    @staticmethod
+    def _build_payload(content: str, model: str, mode: str, img_ids: List[str], img_uris: List[str], is_video: bool = False, post_id: str = None) -> Dict:
+        """构建请求载荷"""
+        # 视频模型特殊处理
+        if is_video and img_uris:
+            img_msg = f"https://grok.com/imagine/{post_id}" if post_id else f"https://assets.grok.com/post/{img_uris[0]}"
+            return {
+                "temporary": True,
+                "modelName": "grok-3",
+                "message": f"{img_msg}  {content} --mode=custom",
+                "fileAttachments": img_ids,
+                "toolOverrides": {"videoGen": True}
+            }
+        
+        # 标准载荷
+        return {
             "temporary": setting.grok_config.get("temporary", True),
-            "modelName": model_name,
+            "modelName": model,
             "message": content,
-            "fileAttachments": image_attachments,
+            "fileAttachments": img_ids,
             "imageAttachments": [],
             "disableSearch": False,
             "enableImageGeneration": True,
@@ -124,125 +183,134 @@ class GrokClient:
             "isReasoning": False,
             "webpageUrls": [],
             "disableTextFollowUps": True,
-            "responseMetadata": {"requestModelDetails": {"modelId": model_name}},
+            "responseMetadata": {"requestModelDetails": {"modelId": model}},
             "disableMemory": False,
             "forceSideBySide": False,
-            "modelMode": model_mode,
+            "modelMode": mode,
             "isAsyncChat": False
         }
-        
-        # 视频模型特殊配置
-        if is_video_model:
-            payload["toolOverrides"] = {"videoGen": True}
-            logger.debug("[Client] 视频模型载荷配置: toolOverrides.videoGen = True")
-        
-        return payload
 
     @staticmethod
-    async def _send_request(payload: dict, auth_token: str, model: str, stream: bool):
-        """发送HTTP请求到Grok API"""
-        # 验证认证令牌
-        if not auth_token:
+    async def _request(payload: dict, token: str, model: str, stream: bool, post_id: str = None):
+        """发送请求"""
+        if not token:
             raise GrokApiException("认证令牌缺失", "NO_AUTH_TOKEN")
 
-        try:
-            # 构建请求头和代理
-            headers = GrokClient._build_headers(auth_token)
-            proxy_config = GrokClient._get_proxy()
-
-            # 构建请求参数
-            request_kwargs = {
-                "headers": headers,
-                "data": json.dumps(payload),
-                "impersonate": IMPERSONATE_BROWSER,
-                "timeout": REQUEST_TIMEOUT,
-                "stream": True,
-                "proxies": proxy_config if proxy_config else None
-            }
-
-            # 在线程池中执行同步HTTP请求，避免阻塞事件循环
-            response = await asyncio.to_thread(
-                curl_requests.post,
-                GROK_API_ENDPOINT,
-                **request_kwargs
-            )
-
-            logger.debug(f"[Client] API响应状态码: {response.status_code}")
-
-            # 处理非成功响应
-            if response.status_code != 200:
-                GrokClient._handle_error(response, auth_token)
-
-            # 请求成功，重置失败计数
-            asyncio.create_task(token_manager.reset_failure(auth_token))
-
-            # 处理并返回响应
-            return await GrokClient._process_response(response, auth_token, model, stream)
-
-        except curl_requests.RequestsError as e:
-            logger.error(f"[Client] 网络请求错误: {e}")
-            raise GrokApiException(f"网络错误: {e}", "NETWORK_ERROR") from e
-        except json.JSONDecodeError as e:
-            logger.error(f"[Client] JSON解析错误: {e}")
-            raise GrokApiException(f"JSON解析错误: {e}", "JSON_ERROR") from e
-        except Exception as e:
-            logger.error(f"[Client] 未知请求错误: {type(e).__name__}: {e}")
-            raise GrokApiException(f"请求处理错误: {e}", "REQUEST_ERROR") from e
+        # 403重试配置
+        max_retries = 5
+        retry_count = 0
+        
+        while retry_count <= max_retries:
+            try:
+                # 构建请求
+                headers = GrokClient._build_headers(token)
+                if model == "grok-imagine-0.9":
+                    file_attachments = payload.get("fileAttachments", [])
+                    ref_id = post_id or (file_attachments[0] if file_attachments else "")
+                    if ref_id:
+                        headers["Referer"] = f"https://grok.com/imagine/{ref_id}"
+                
+                # 异步获取代理
+                from app.core.proxy_pool import proxy_pool
+                
+                # 如果是重试且使用代理池，强制刷新代理
+                if retry_count > 0 and proxy_pool._enabled:
+                    logger.info(f"[Client] 403重试 {retry_count}/{max_retries}，刷新代理...")
+                    proxy = await proxy_pool.force_refresh()
+                else:
+                    proxy = await setting.get_proxy_async("service")
+                
+                proxies = {"http": proxy, "https": proxy} if proxy else None
+                
+                # 执行请求
+                response = await asyncio.to_thread(
+                    curl_requests.post,
+                    API_ENDPOINT,
+                    headers=headers,
+                    data=orjson.dumps(payload),
+                    impersonate=BROWSER,
+                    timeout=TIMEOUT,
+                    stream=True,
+                    proxies=proxies
+                )
+                
+                # 内层403重试：仅当有代理池时触发
+                if response.status_code == 403 and proxy_pool._enabled:
+                    retry_count += 1
+                    
+                    if retry_count <= max_retries:
+                        logger.warning(f"[Client] 遇到403错误，正在重试 ({retry_count}/{max_retries})...")
+                        await asyncio.sleep(0.5)
+                        continue
+                    
+                    # 内层重试全部失败
+                    logger.error(f"[Client] 403错误，已重试{retry_count-1}次，放弃")
+                
+                # 检查响应状态
+                if response.status_code != 200:
+                    GrokClient._handle_error(response, token)
+                
+                # 成功 - 重置失败计数
+                asyncio.create_task(token_manager.reset_failure(token))
+                
+                # 如果是重试成功，记录日志
+                if retry_count > 0:
+                    logger.info(f"[Client] 重试成功！第{retry_count}次重试后请求成功")
+                
+                # 处理响应
+                result = (GrokResponseProcessor.process_stream(response, token) if stream 
+                         else await GrokResponseProcessor.process_normal(response, token, model))
+                
+                asyncio.create_task(GrokClient._update_limits(token, model))
+                return result
+                
+            except curl_requests.RequestsError as e:
+                logger.error(f"[Client] 网络错误: {e}")
+                raise GrokApiException(f"网络错误: {e}", "NETWORK_ERROR") from e
+            except GrokApiException:
+                # 重新抛出GrokApiException（包括403错误）
+                raise
+            except Exception as e:
+                logger.error(f"[Client] 请求错误: {e}")
+                raise GrokApiException(f"请求错误: {e}", "REQUEST_ERROR") from e
+        
+        # 理论上不应该到这里，但以防万一
+        raise GrokApiException("403错误：已达到最大重试次数", "MAX_RETRIES_EXCEEDED")
 
     @staticmethod
-    def _build_headers(auth_token: str) -> Dict[str, str]:
+    def _build_headers(token: str) -> Dict[str, str]:
         """构建请求头"""
         headers = get_dynamic_headers("/rest/app-chat/conversations/new")
-
-        # 构建Cookie
-        cf_clearance = setting.grok_config.get("cf_clearance", "")
-        headers["Cookie"] = f"{auth_token};{cf_clearance}" if cf_clearance else auth_token
-
+        cf = setting.grok_config.get("cf_clearance", "")
+        headers["Cookie"] = f"{token};{cf}" if cf else token
         return headers
 
     @staticmethod
-    def _get_proxy() -> Dict[str, str]:
-        """获取代理配置"""
-        proxy_url = setting.grok_config.get("proxy_url", "")
-        if proxy_url:
-            return {"http": proxy_url, "https": proxy_url}
-        return {}
-
-    @staticmethod
-    def _handle_error(response, auth_token: str):
-        """处理错误响应"""
-        try:
-            error_data = response.json()
-            error_message = str(error_data)
-        except Exception as e:
-            error_data = response.text
-            error_message = error_data[:200] if error_data else e
-
-        # 记录Token失败
-        asyncio.create_task(token_manager.record_failure(auth_token, response.status_code, error_message))
-
+    def _handle_error(response, token: str):
+        """处理错误"""
+        if response.status_code == 403:
+            msg = "您的IP被拦截，请尝试以下方法之一: 1.更换IP 2.使用代理 3.配置CF值"
+            data = {"cf_blocked": True, "status": 403}
+            logger.warning(f"[Client] {msg}")
+        else:
+            try:
+                data = response.json()
+                msg = str(data)
+            except:
+                data = response.text
+                msg = data[:200] if data else "未知错误"
+        
+        asyncio.create_task(token_manager.record_failure(token, response.status_code, msg))
         raise GrokApiException(
-            f"请求失败: {response.status_code} - {error_message}",
+            f"请求失败: {response.status_code} - {msg}",
             "HTTP_ERROR",
-            {"status": response.status_code, "data": error_data}
+            {"status": response.status_code, "data": data}
         )
 
     @staticmethod
-    async def _process_response(response, auth_token: str, model: str, stream: bool):
-        """处理API响应"""
-        if stream:
-            result = GrokResponseProcessor.process_stream(response, auth_token)
-            asyncio.create_task(GrokClient._update_rate_limits(auth_token, model))
-        else:
-            result = await GrokResponseProcessor.process_normal(response, auth_token, model)
-            asyncio.create_task(GrokClient._update_rate_limits(auth_token, model))
-
-        return result
-
-    @staticmethod
-    async def _update_rate_limits(auth_token: str, model: str):
-        """异步更新速率限制信息"""
+    async def _update_limits(token: str, model: str):
+        """更新速率限制"""
         try:
-            await token_manager.check_limits(auth_token, model)
+            await token_manager.check_limits(token, model)
         except Exception as e:
-            logger.error(f"[Client] 更新速率限制失败: {e}")
\ No newline at end of file
+            logger.error(f"[Client] 更新限制失败: {e}")
\ No newline at end of file
diff --git a/app/services/grok/create.py b/app/services/grok/create.py
new file mode 100644
index 0000000..3652bc3
--- /dev/null
+++ b/app/services/grok/create.py
@@ -0,0 +1,140 @@
+"""Post创建管理器 - 用于视频生成前的会话创建"""
+
+import asyncio
+import orjson
+from typing import Dict, Any, Optional
+from curl_cffi.requests import AsyncSession
+
+from app.services.grok.statsig import get_dynamic_headers
+from app.core.exception import GrokApiException
+from app.core.config import setting
+from app.core.logger import logger
+
+
+# 常量
+ENDPOINT = "https://grok.com/rest/media/post/create"
+TIMEOUT = 30
+BROWSER = "chrome133a"
+
+
+class PostCreateManager:
+    """会话创建管理器"""
+
+    @staticmethod
+    async def create(file_id: str, file_uri: str, auth_token: str) -> Optional[Dict[str, Any]]:
+        """创建会话记录
+        
+        Args:
+            file_id: 文件ID
+            file_uri: 文件URI
+            auth_token: 认证令牌
+            
+        Returns:
+            会话信息字典，包含post_id等
+        """
+        # 参数验证
+        if not file_id or not file_uri:
+            raise GrokApiException("文件ID或URI缺失", "INVALID_PARAMS")
+        if not auth_token:
+            raise GrokApiException("认证令牌缺失", "NO_AUTH_TOKEN")
+
+        try:
+            # 构建请求
+            data = {
+                "media_url": f"https://assets.grok.com/{file_uri}",
+                "media_type": "MEDIA_POST_TYPE_IMAGE"
+            }
+            
+            cf = setting.grok_config.get("cf_clearance", "")
+            headers = {
+                **get_dynamic_headers("/rest/media/post/create"),
+                "Cookie": f"{auth_token};{cf}" if cf else auth_token
+            }
+            
+            # 外层重试：可配置状态码（401/429等）
+            retry_codes = setting.grok_config.get("retry_status_codes", [401, 429])
+            MAX_OUTER_RETRY = 3
+            
+            for outer_retry in range(MAX_OUTER_RETRY + 1):  # +1 确保实际重试3次
+                # 内层重试：403代理池重试
+                max_403_retries = 5
+                retry_403_count = 0
+                
+                while retry_403_count <= max_403_retries:
+                    # 异步获取代理（支持代理池）
+                    from app.core.proxy_pool import proxy_pool
+                    
+                    # 如果是403重试且使用代理池，强制刷新代理
+                    if retry_403_count > 0 and proxy_pool._enabled:
+                        logger.info(f"[PostCreate] 403重试 {retry_403_count}/{max_403_retries}，刷新代理...")
+                        proxy = await proxy_pool.force_refresh()
+                    else:
+                        proxy = await setting.get_proxy_async("service")
+                    
+                    proxies = {"http": proxy, "https": proxy} if proxy else None
+
+                    # 发送请求
+                    async with AsyncSession() as session:
+                        response = await session.post(
+                            ENDPOINT,
+                            headers=headers,
+                            json=data,
+                            impersonate=BROWSER,
+                            timeout=TIMEOUT,
+                            proxies=proxies
+                        )
+
+                        # 内层403重试：仅当有代理池时触发
+                        if response.status_code == 403 and proxy_pool._enabled:
+                            retry_403_count += 1
+                            
+                            if retry_403_count <= max_403_retries:
+                                logger.warning(f"[PostCreate] 遇到403错误，正在重试 ({retry_403_count}/{max_403_retries})...")
+                                await asyncio.sleep(0.5)
+                                continue
+                            
+                            # 内层重试全部失败
+                            logger.error(f"[PostCreate] 403错误，已重试{retry_403_count-1}次，放弃")
+                        
+                        # 检查可配置状态码错误 - 外层重试
+                        if response.status_code in retry_codes:
+                            if outer_retry < MAX_OUTER_RETRY:
+                                delay = (outer_retry + 1) * 0.1  # 渐进延迟：0.1s, 0.2s, 0.3s
+                                logger.warning(f"[PostCreate] 遇到{response.status_code}错误，外层重试 ({outer_retry+1}/{MAX_OUTER_RETRY})，等待{delay}s...")
+                                await asyncio.sleep(delay)
+                                break  # 跳出内层循环，进入外层重试
+                            else:
+                                logger.error(f"[PostCreate] {response.status_code}错误，已重试{outer_retry}次，放弃")
+                                raise GrokApiException(f"创建失败: {response.status_code}错误", "CREATE_ERROR")
+
+                        if response.status_code == 200:
+                            result = response.json()
+                            post_id = result.get("post", {}).get("id", "")
+                            
+                            if outer_retry > 0 or retry_403_count > 0:
+                                logger.info(f"[PostCreate] 重试成功！")
+                            
+                            logger.debug(f"[PostCreate] 成功，会话ID: {post_id}")
+                            return {
+                                "post_id": post_id,
+                                "file_id": file_id,
+                                "file_uri": file_uri,
+                                "success": True,
+                                "data": result
+                            }
+                        
+                        # 其他错误处理
+                        try:
+                            error = response.json()
+                            msg = f"状态码: {response.status_code}, 详情: {error}"
+                        except:
+                            msg = f"状态码: {response.status_code}, 详情: {response.text[:200]}"
+                        
+                        logger.error(f"[PostCreate] 失败: {msg}")
+                        raise GrokApiException(f"创建失败: {msg}", "CREATE_ERROR")
+
+        except GrokApiException:
+            raise
+        except Exception as e:
+            logger.error(f"[PostCreate] 异常: {e}")
+            raise GrokApiException(f"创建异常: {e}", "CREATE_ERROR") from e
diff --git a/app/services/grok/processer.py b/app/services/grok/processer.py
index f671a55..2c51c69 100644
--- a/app/services/grok/processer.py
+++ b/app/services/grok/processer.py
@@ -1,10 +1,10 @@
-"""Grok API 响应处理器模块"""
+"""Grok API 响应处理器 - 处理流式和非流式响应"""
 
-import json
+import orjson
 import uuid
 import time
 import asyncio
-from typing import AsyncGenerator
+from typing import AsyncGenerator, Tuple
 
 from app.core.config import setting
 from app.core.exception import GrokApiException
@@ -21,61 +21,43 @@ from app.services.grok.cache import image_cache_service, video_cache_service
 
 
 class StreamTimeoutManager:
-    """流式响应超时管理器"""
+    """流式响应超时管理"""
     
-    def __init__(self, chunk_timeout: int = 120, first_response_timeout: int = 30, total_timeout: int = 600):
-        """初始化超时管理器
-        
-        Args:
-            chunk_timeout: 数据块间隔超时（秒）
-            first_response_timeout: 首次响应超时（秒）
-            total_timeout: 总超时限制（秒，0表示不限制）
-        """
+    def __init__(self, chunk_timeout: int = 120, first_timeout: int = 30, total_timeout: int = 600):
         self.chunk_timeout = chunk_timeout
-        self.first_response_timeout = first_response_timeout
+        self.first_timeout = first_timeout
         self.total_timeout = total_timeout
-        
         self.start_time = asyncio.get_event_loop().time()
         self.last_chunk_time = self.start_time
-        self.first_chunk_received = False
+        self.first_received = False
     
-    def check_timeout(self) -> tuple[bool, str]:
-        """检查是否超时
+    def check_timeout(self) -> Tuple[bool, str]:
+        """检查超时"""
+        now = asyncio.get_event_loop().time()
         
-        Returns:
-            (is_timeout, timeout_message): 是否超时及超时信息
-        """
-        current_time = asyncio.get_event_loop().time()
+        if not self.first_received and now - self.start_time > self.first_timeout:
+            return True, f"首次响应超时({self.first_timeout}秒)"
         
-        # 检查首次响应超时
-        if not self.first_chunk_received:
-            if current_time - self.start_time > self.first_response_timeout:
-                return True, f"首次响应超时 ({self.first_response_timeout}秒未收到首个数据块)"
+        if self.total_timeout > 0 and now - self.start_time > self.total_timeout:
+            return True, f"总超时({self.total_timeout}秒)"
         
-        # 检查总超时
-        if self.total_timeout > 0:
-            if current_time - self.start_time > self.total_timeout:
-                return True, f"流式响应总超时 ({self.total_timeout}秒)"
-        
-        # 检查数据块间隔超时
-        if self.first_chunk_received:
-            if current_time - self.last_chunk_time > self.chunk_timeout:
-                return True, f"数据块间隔超时 ({self.chunk_timeout}秒无新数据)"
+        if self.first_received and now - self.last_chunk_time > self.chunk_timeout:
+            return True, f"数据块超时({self.chunk_timeout}秒)"
         
         return False, ""
     
-    def mark_chunk_received(self):
-        """标记收到数据块"""
+    def mark_received(self):
+        """标记收到数据"""
         self.last_chunk_time = asyncio.get_event_loop().time()
-        self.first_chunk_received = True
+        self.first_received = True
     
-    def get_total_duration(self) -> float:
-        """获取总耗时（秒）"""
+    def duration(self) -> float:
+        """获取总耗时"""
         return asyncio.get_event_loop().time() - self.start_time
 
 
 class GrokResponseProcessor:
-    """Grok API 响应处理器"""
+    """Grok响应处理器"""
 
     @staticmethod
     async def process_normal(response, auth_token: str, model: str = None) -> OpenAIChatCompletionResponse:
@@ -86,7 +68,7 @@ class GrokResponseProcessor:
                 if not chunk:
                     continue
 
-                data = json.loads(chunk.decode("utf-8"))
+                data = orjson.loads(chunk)
 
                 # 错误检查
                 if error := data.get("error"):
@@ -96,312 +78,190 @@ class GrokResponseProcessor:
                         {"code": error.get("code")}
                     )
 
-                # 提取响应数据
                 grok_resp = data.get("result", {}).get("response", {})
                 
-                # 提取视频数据
+                # 视频响应
                 if video_resp := grok_resp.get("streamingVideoGenerationResponse"):
                     if video_url := video_resp.get("videoUrl"):
-                        logger.debug(f"[Processor] 检测到视频生成: {video_url}")
-                        full_video_url = f"https://assets.grok.com/{video_url}"
-                        
-                        # 下载并缓存视频
-                        try:
-                            cache_path = await video_cache_service.download_video(f"/{video_url}", auth_token)
-                            if cache_path:
-                                video_path = video_url.replace('/', '-')
-                                base_url = setting.global_config.get("base_url", "")
-                                local_video_url = f"{base_url}/images/{video_path}" if base_url else f"/images/{video_path}"
-                                content = f'<video src="{local_video_url}" controls="controls" width="500" height="300"></video>\n'
-                            else:
-                                content = f'<video src="{full_video_url}" controls="controls" width="500" height="300"></video>\n'
-                        except Exception as e:
-                            logger.warning(f"[Processor] 缓存视频失败: {e}")
-                            content = f'<video src="{full_video_url}" controls="controls" width="500" height="300"></video>\n'
-                        
-                        # 返回视频响应
-                        result = OpenAIChatCompletionResponse(
-                            id=f"chatcmpl-{uuid.uuid4()}",
-                            object="chat.completion",
-                            created=int(time.time()),
-                            model=model or "grok-imagine-0.9",
-                            choices=[OpenAIChatCompletionChoice(
-                                index=0,
-                                message=OpenAIChatCompletionMessage(
-                                    role="assistant",
-                                    content=content
-                                ),
-                                finish_reason="stop"
-                            )],
-                            usage=None
-                        )
+                        content = await GrokResponseProcessor._build_video_content(video_url, auth_token)
+                        result = GrokResponseProcessor._build_response(content, model or "grok-imagine-0.9")
                         response_closed = True
                         response.close()
                         return result
 
-                # 提取模型响应
+                # 模型响应
                 model_response = grok_resp.get("modelResponse")
                 if not model_response:
                     continue
 
-                # 检查 modelResponse 中的错误
                 if error_msg := model_response.get("error"):
-                    raise GrokApiException(
-                        f"模型响应错误: {error_msg}",
-                        "MODEL_ERROR"
-                    )
+                    raise GrokApiException(f"模型错误: {error_msg}", "MODEL_ERROR")
 
-                # 构建响应内容
-                model_name = model_response.get("model")
+                # 构建内容
                 content = model_response.get("message", "")
+                model_name = model_response.get("model")
 
-                # 提取图片数据
+                # 处理图片
                 if images := model_response.get("generatedImageUrls"):
-                    # 获取图片返回模式
-                    image_mode = setting.global_config.get("image_mode", "url")
+                    content = await GrokResponseProcessor._append_images(content, images, auth_token)
 
-                    for img in images:
-                        try:
-                            if image_mode == "base64":
-                                # base64 模式：下载并转换为 base64
-                                base64_str = await image_cache_service.download_base64(f"/{img}", auth_token)
-                                if base64_str:
-                                    content += f"\n![Generated Image]({base64_str})"
-                                else:
-                                    content += f"\n![Generated Image](https://assets.grok.com/{img})"
-                            else:
-                                # url 模式：缓存并返回链接
-                                cache_path = await image_cache_service.download_image(f"/{img}", auth_token)
-                                if cache_path:
-                                    img_path = img.replace('/', '-')
-                                    base_url = setting.global_config.get("base_url", "")
-                                    img_url = f"{base_url}/images/{img_path}" if base_url else f"/images/{img_path}"
-                                    content += f"\n![Generated Image]({img_url})"
-                                else:
-                                    content += f"\n![Generated Image](https://assets.grok.com/{img})"
-                        except Exception as e:
-                            logger.warning(f"[Processor] 处理图片失败: {e}")
-                            content += f"\n![Generated Image](https://assets.grok.com/{img})"
-
-                # 返回 OpenAI 响应格式
-                result = OpenAIChatCompletionResponse(
-                    id=f"chatcmpl-{uuid.uuid4()}",
-                    object="chat.completion",
-                    created=int(time.time()),
-                    model=model_name,
-                    choices=[OpenAIChatCompletionChoice(
-                        index=0,
-                        message=OpenAIChatCompletionMessage(
-                            role="assistant",
-                            content=content
-                        ),
-                        finish_reason="stop"
-                    )],
-                    usage=None
-                )
+                result = GrokResponseProcessor._build_response(content, model_name)
                 response_closed = True
                 response.close()
                 return result
 
             raise GrokApiException("无响应数据", "NO_RESPONSE")
 
-        except json.JSONDecodeError as e:
+        except orjson.JSONDecodeError as e:
             logger.error(f"[Processor] JSON解析失败: {e}")
             raise GrokApiException(f"JSON解析失败: {e}", "JSON_ERROR") from e
         except Exception as e:
-            logger.error(f"[Processor] 处理响应时发生未知错误: {type(e).__name__}: {e}")
+            logger.error(f"[Processor] 处理错误: {type(e).__name__}: {e}")
             raise GrokApiException(f"响应处理错误: {e}", "PROCESS_ERROR") from e
         finally:
-            # 确保响应对象被关闭，避免双重释放
             if not response_closed and hasattr(response, 'close'):
                 try:
                     response.close()
                 except Exception as e:
-                    logger.warning(f"[Processor] 关闭响应对象时出错: {e}")
+                    logger.warning(f"[Processor] 关闭响应失败: {e}")
 
     @staticmethod
     async def process_stream(response, auth_token: str) -> AsyncGenerator[str, None]:
         """处理流式响应"""
-        # 流式生成状态
+        # 状态变量
         is_image = False
         is_thinking = False
         thinking_finished = False
-        chunk_index = 0
         model = None
         filtered_tags = setting.grok_config.get("filtered_tags", "").split(",")
         video_progress_started = False
         last_video_progress = -1
         response_closed = False
+        show_thinking = setting.grok_config.get("show_thinking", True)
 
-        # 初始化超时管理器
-        timeout_manager = StreamTimeoutManager(
+        # 超时管理
+        timeout_mgr = StreamTimeoutManager(
             chunk_timeout=setting.grok_config.get("stream_chunk_timeout", 120),
-            first_response_timeout=setting.grok_config.get("stream_first_response_timeout", 30),
+            first_timeout=setting.grok_config.get("stream_first_response_timeout", 30),
             total_timeout=setting.grok_config.get("stream_total_timeout", 600)
         )
 
-        def make_chunk(chunk_content: str, finish: str = None):
-            """生成OpenAI格式的响应块"""
+        def make_chunk(content: str, finish: str = None):
+            """生成响应块"""
             chunk_data = OpenAIChatCompletionChunkResponse(
                 id=f"chatcmpl-{uuid.uuid4()}",
                 created=int(time.time()),
                 model=model or "grok-4-mini-thinking-tahoe",
                 choices=[OpenAIChatCompletionChunkChoice(
-                    index=chunk_index,
+                    index=0,
                     delta=OpenAIChatCompletionChunkMessage(
                         role="assistant",
-                        content=chunk_content
-                    ) if chunk_content else {},
+                        content=content
+                    ) if content else {},
                     finish_reason=finish
                 )]
-            ).model_dump()
-            # SSE 格式返回
-            return f"data: {json.dumps(chunk_data)}\n\n"
+            )
+            return f"data: {chunk_data.model_dump_json()}\n\n"
 
         try:
             for chunk in response.iter_lines():
                 # 超时检查
-                is_timeout, timeout_msg = timeout_manager.check_timeout()
+                is_timeout, timeout_msg = timeout_mgr.check_timeout()
                 if is_timeout:
                     logger.warning(f"[Processor] {timeout_msg}")
                     yield make_chunk("", "stop")
                     yield "data: [DONE]\n\n"
                     return
 
-                logger.debug(f"[Processor] 接收到数据块: {len(chunk)} bytes")
+                logger.debug(f"[Processor] 收到数据块: {len(chunk)} bytes")
                 if not chunk:
                     continue
 
                 try:
-                    data = json.loads(chunk.decode("utf-8"))
+                    data = orjson.loads(chunk)
 
                     # 错误检查
                     if error := data.get("error"):
                         error_msg = error.get('message', '未知错误')
-                        logger.error(f"[Processor] Grok API返回错误: {error_msg}")
+                        logger.error(f"[Processor] API错误: {error_msg}")
                         yield make_chunk(f"Error: {error_msg}", "stop")
                         yield "data: [DONE]\n\n"
                         return
 
-                    # 提取响应数据
                     grok_resp = data.get("result", {}).get("response", {})
-                    logger.debug(f"[Processor] 解析响应数据: {len(grok_resp)} 字段")
+                    logger.debug(f"[Processor] 解析响应: {len(grok_resp)} bytes")
                     if not grok_resp:
                         continue
+                    
+                    timeout_mgr.mark_received()
 
-                    # 更新模型名称
+                    # 更新模型
                     if user_resp := grok_resp.get("userResponse"):
                         if m := user_resp.get("model"):
                             model = m
 
-                    # 提取视频数据
+                    # 视频处理
                     if video_resp := grok_resp.get("streamingVideoGenerationResponse"):
                         progress = video_resp.get("progress", 0)
+                        v_url = video_resp.get("videoUrl")
                         
+                        # 进度更新
                         if progress > last_video_progress:
                             last_video_progress = progress
-                            
-                            # 添加 <think> 标签
-                            if not video_progress_started:
-                                content = f"<think>视频已生成{progress}%\n"
-                                video_progress_started = True
-                            elif progress < 100:
-                                content = f"视频已生成{progress}%\n"
-                            else:
-                                # 进度100%时关闭 <think> 标签并立即处理视频
-                                content = f"视频已生成{progress}%</think>\n"
-                                
-                                # 立即下载并缓存视频
-                                if v_url := video_resp.get("videoUrl"):
-                                    logger.debug(f"[Processor] 视频生成完成: {v_url}")
-                                    full_video_url = f"https://assets.grok.com/{v_url}"
-                                    
-                                    try:
-                                        cache_path = await video_cache_service.download_video(f"/{v_url}", auth_token)
-                                        if cache_path:
-                                            video_path = v_url.replace('/', '-')
-                                            base_url = setting.global_config.get("base_url", "")
-                                            local_video_url = f"{base_url}/images/{video_path}" if base_url else f"/images/{video_path}"
-                                            content += f'<video src="{local_video_url}" controls="controls"></video>\n'
-                                        else:
-                                            content += f'<video src="{full_video_url}" controls="controls"></video>\n'
-                                    except Exception as e:
-                                        logger.warning(f"[Processor] 缓存视频失败: {e}")
-                                        content += f'<video src="{full_video_url}" controls="controls"></video>\n'
-
-                            yield make_chunk(content)
-                            timeout_manager.mark_chunk_received()
-                            chunk_index += 1
+                            if show_thinking:
+                                if not video_progress_started:
+                                    content = f"<think>视频已生成{progress}%\n"
+                                    video_progress_started = True
+                                elif progress < 100:
+                                    content = f"视频已生成{progress}%\n"
+                                else:
+                                    content = f"视频已生成{progress}%</think>\n"
+                                yield make_chunk(content)
+                        
+                        # 视频URL
+                        if v_url:
+                            logger.debug("[Processor] 视频生成完成")
+                            video_content = await GrokResponseProcessor._build_video_content(v_url, auth_token)
+                            yield make_chunk(video_content)
                         
                         continue
 
-                    # 检查生成模式
+                    # 图片模式
                     if grok_resp.get("imageAttachmentInfo"):
                         is_image = True
 
-                    # 获取token
                     token = grok_resp.get("token", "")
 
-                    # 提取图片数据
+                    # 图片处理
                     if is_image:
                         if model_resp := grok_resp.get("modelResponse"):
-                            # 获取图片返回模式
                             image_mode = setting.global_config.get("image_mode", "url")
-
-                            # 初始化内容变量
                             content = ""
 
-                            # 生成图片链接并缓存
                             for img in model_resp.get("generatedImageUrls", []):
                                 try:
                                     if image_mode == "base64":
-                                        # base64 模式：下载并转换为 base64
+                                        # Base64模式 - 分块发送
                                         base64_str = await image_cache_service.download_base64(f"/{img}", auth_token)
                                         if base64_str:
-                                            # 分块发送 base64 数据，每 8KB 一个 chunk
-                                            markdown_prefix = "![Generated Image](data:"
-                                            markdown_suffix = ")\n"
-
-                                            # 提取 data URL 的 mime 和 base64 部分
-                                            if base64_str.startswith("data:"):
+                                            # 分块发送大数据
+                                            if not base64_str.startswith("data:"):
                                                 parts = base64_str.split(",", 1)
                                                 if len(parts) == 2:
-                                                    mime_part = parts[0] + ","
-                                                    b64_data = parts[1]
-
-                                                    # 发送前缀
-                                                    yield make_chunk(markdown_prefix + mime_part)
-                                                    timeout_manager.mark_chunk_received()
-                                                    chunk_index += 1
-
-                                                    # 分块发送 base64 数据
-                                                    chunk_size = 8192
-                                                    for i in range(0, len(b64_data), chunk_size):
-                                                        chunk_data = b64_data[i:i + chunk_size]
-                                                        yield make_chunk(chunk_data)
-                                                        timeout_manager.mark_chunk_received()
-                                                        chunk_index += 1
-
-                                                    # 发送后缀
-                                                    yield make_chunk(markdown_suffix)
-                                                    timeout_manager.mark_chunk_received()
-                                                    chunk_index += 1
+                                                    yield make_chunk(f"![Generated Image](data:{parts[0]},")
+                                                    # 8KB分块
+                                                    for i in range(0, len(parts[1]), 8192):
+                                                        yield make_chunk(parts[1][i:i+8192])
+                                                    yield make_chunk(")\n")
                                                 else:
                                                     yield make_chunk(f"![Generated Image]({base64_str})\n")
-                                                    timeout_manager.mark_chunk_received()
-                                                    chunk_index += 1
                                             else:
                                                 yield make_chunk(f"![Generated Image]({base64_str})\n")
-                                                timeout_manager.mark_chunk_received()
-                                                chunk_index += 1
                                         else:
                                             yield make_chunk(f"![Generated Image](https://assets.grok.com/{img})\n")
-                                            timeout_manager.mark_chunk_received()
-                                            chunk_index += 1
                                     else:
-                                        # url 模式：缓存并返回链接
+                                        # URL模式
                                         await image_cache_service.download_image(f"/{img}", auth_token)
-                                        # 本地图片路径
                                         img_path = img.replace('/', '-')
                                         base_url = setting.global_config.get("base_url", "")
                                         img_url = f"{base_url}/images/{img_path}" if base_url else f"/images/{img_path}"
@@ -410,97 +270,154 @@ class GrokResponseProcessor:
                                     logger.warning(f"[Processor] 处理图片失败: {e}")
                                     content += f"![Generated Image](https://assets.grok.com/{img})\n"
 
-                            # 发送内容
                             yield make_chunk(content.strip(), "stop")
-                            timeout_manager.mark_chunk_received()
                             return
                         elif token:
                             yield make_chunk(token)
-                            timeout_manager.mark_chunk_received()
-                            chunk_index += 1
 
-                    # 提取对话数据
+                    # 对话处理
                     else:
-                        # 过滤 list 格式的 token
                         if isinstance(token, list):
                             continue
 
-                        # 过滤特定标签
                         if any(tag in token for tag in filtered_tags if token):
                             continue
 
-                        # 获取当前状态
                         current_is_thinking = grok_resp.get("isThinking", False)
                         message_tag = grok_resp.get("messageTag")
 
-                        # 跳过后续的 <think> 标签
                         if thinking_finished and current_is_thinking:
                             continue
 
-                        # 检查 toolUsageCardId
+                        # 搜索结果处理
                         if grok_resp.get("toolUsageCardId"):
                             if web_search := grok_resp.get("webSearchResults"):
                                 if current_is_thinking:
-                                    # 封装搜索结果
-                                    for result in web_search.get("results", []):
-                                        title = result.get("title", "")
-                                        url = result.get("url", "")
-                                        preview = result.get("preview", "")
-                                        preview_clean = preview.replace("\n", "") if isinstance(preview, str) else ""
-                                        token += f'\n- [{title}]({url} "{preview_clean}")'
-                                    token += "\n"
+                                    if show_thinking:
+                                        for result in web_search.get("results", []):
+                                            title = result.get("title", "")
+                                            url = result.get("url", "")
+                                            preview = result.get("preview", "")
+                                            preview_clean = preview.replace("\n", "") if isinstance(preview, str) else ""
+                                            token += f'\n- [{title}]({url} "{preview_clean}")'
+                                        token += "\n"
+                                    else:
+                                        continue
                                 else:
-                                    # 有 webSearchResults 但 isThinking 为 false
                                     continue
                             else:
-                                # 没有 webSearchResults
                                 continue
 
                         if token:
                             content = token
 
-                            # header 在 token 后换行
                             if message_tag == "header":
                                 content = f"\n\n{token}\n\n"
 
-                            # is_thinking 状态切换
+                            # Thinking状态切换
+                            should_skip = False
                             if not is_thinking and current_is_thinking:
-                                content = f"<think>\n{content}"
+                                if show_thinking:
+                                    content = f"<think>\n{content}"
+                                else:
+                                    should_skip = True
                             elif is_thinking and not current_is_thinking:
-                                content = f"\n</think>\n{content}"
+                                if show_thinking:
+                                    content = f"\n</think>\n{content}"
                                 thinking_finished = True
+                            elif current_is_thinking:
+                                if not show_thinking:
+                                    should_skip = True
 
-                            yield make_chunk(content)
-                            timeout_manager.mark_chunk_received()
-                            chunk_index += 1
+                            if not should_skip:
+                                yield make_chunk(content)
+                            
                             is_thinking = current_is_thinking
 
-                except (json.JSONDecodeError, UnicodeDecodeError) as e:
-                    logger.warning(f"[Processor] 解析chunk失败: {e}")
+                except (orjson.JSONDecodeError, UnicodeDecodeError) as e:
+                    logger.warning(f"[Processor] 解析失败: {e}")
                     continue
                 except Exception as e:
-                    logger.warning(f"[Processor] 处理chunk出错: {e}")
+                    logger.warning(f"[Processor] 处理出错: {e}")
                     continue
 
-            # 发送结束块
             yield make_chunk("", "stop")
-
-            # 发送流结束标记
             yield "data: [DONE]\n\n"
-            
-            # 记录流式响应统计
-            logger.info(f"[Processor] 流式响应完成，总耗时: {timeout_manager.get_total_duration():.2f}秒")
+            logger.info(f"[Processor] 流式完成，耗时: {timeout_mgr.duration():.2f}秒")
 
         except Exception as e:
-            logger.error(f"[Processor] 流式处理严重错误: {e}")
+            logger.error(f"[Processor] 严重错误: {e}")
             yield make_chunk(f"处理错误: {e}", "error")
-            # 发送流结束标记
             yield "data: [DONE]\n\n"
         finally:
-            # 确保响应对象被关闭
             if not response_closed and hasattr(response, 'close'):
                 try:
                     response.close()
-                    logger.debug("[Processor] 流式响应对象已关闭")
+                    logger.debug("[Processor] 响应已关闭")
                 except Exception as e:
-                    logger.warning(f"[Processor] 关闭流式响应对象时出错: {e}")
\ No newline at end of file
+                    logger.warning(f"[Processor] 关闭失败: {e}")
+
+    @staticmethod
+    async def _build_video_content(video_url: str, auth_token: str) -> str:
+        """构建视频内容"""
+        logger.debug(f"[Processor] 检测到视频: {video_url}")
+        full_url = f"https://assets.grok.com/{video_url}"
+        
+        try:
+            cache_path = await video_cache_service.download_video(f"/{video_url}", auth_token)
+            if cache_path:
+                video_path = video_url.replace('/', '-')
+                base_url = setting.global_config.get("base_url", "")
+                local_url = f"{base_url}/images/{video_path}" if base_url else f"/images/{video_path}"
+                return f'<video src="{local_url}" controls="controls" width="500" height="300"></video>\n'
+        except Exception as e:
+            logger.warning(f"[Processor] 缓存视频失败: {e}")
+        
+        return f'<video src="{full_url}" controls="controls" width="500" height="300"></video>\n'
+
+    @staticmethod
+    async def _append_images(content: str, images: list, auth_token: str) -> str:
+        """追加图片到内容"""
+        image_mode = setting.global_config.get("image_mode", "url")
+        
+        for img in images:
+            try:
+                if image_mode == "base64":
+                    base64_str = await image_cache_service.download_base64(f"/{img}", auth_token)
+                    if base64_str:
+                        content += f"\n![Generated Image]({base64_str})"
+                    else:
+                        content += f"\n![Generated Image](https://assets.grok.com/{img})"
+                else:
+                    cache_path = await image_cache_service.download_image(f"/{img}", auth_token)
+                    if cache_path:
+                        img_path = img.replace('/', '-')
+                        base_url = setting.global_config.get("base_url", "")
+                        img_url = f"{base_url}/images/{img_path}" if base_url else f"/images/{img_path}"
+                        content += f"\n![Generated Image]({img_url})"
+                    else:
+                        content += f"\n![Generated Image](https://assets.grok.com/{img})"
+            except Exception as e:
+                logger.warning(f"[Processor] 处理图片失败: {e}")
+                content += f"\n![Generated Image](https://assets.grok.com/{img})"
+        
+        return content
+
+    @staticmethod
+    def _build_response(content: str, model: str) -> OpenAIChatCompletionResponse:
+        """构建响应对象"""
+        return OpenAIChatCompletionResponse(
+            id=f"chatcmpl-{uuid.uuid4()}",
+            object="chat.completion",
+            created=int(time.time()),
+            model=model,
+            choices=[OpenAIChatCompletionChoice(
+                index=0,
+                message=OpenAIChatCompletionMessage(
+                    role="assistant",
+                    content=content
+                ),
+                finish_reason="stop"
+            )],
+            usage=None
+        )
diff --git a/app/services/grok/statsig.py b/app/services/grok/statsig.py
index 2c30601..a3d7017 100644
--- a/app/services/grok/statsig.py
+++ b/app/services/grok/statsig.py
@@ -1,44 +1,80 @@
-"""Grok 请求头管理模块"""
+"""Grok 请求头管理 - 生成动态请求头和Statsig ID"""
 
+import base64
+import random
+import string
 import uuid
 from typing import Dict
 
+from app.core.logger import logger
 from app.core.config import setting
 
 
+# 基础请求头
+BASE_HEADERS = {
+    "Accept": "*/*",
+    "Accept-Language": "zh-CN,zh;q=0.9",
+    "Accept-Encoding": "gzip, deflate, br, zstd",
+    "Connection": "keep-alive",
+    "Origin": "https://grok.com",
+    "Priority": "u=1, i",
+    "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/133.0.0.0 Safari/537.36",
+    "Sec-Ch-Ua": '"Not(A:Brand";v="99", "Google Chrome";v="133", "Chromium";v="133"',
+    "Sec-Ch-Ua-Mobile": "?0",
+    "Sec-Ch-Ua-Platform": '"macOS"',
+    "Sec-Fetch-Dest": "empty",
+    "Sec-Fetch-Mode": "cors",
+    "Sec-Fetch-Site": "same-origin",
+    "Baggage": "sentry-environment=production,sentry-public_key=b311e0f2690c81f25e2c4cf6d4f7ce1c",
+}
+
+
+def _random_string(length: int, letters_only: bool = True) -> str:
+    """生成随机字符串"""
+    chars = string.ascii_lowercase if letters_only else string.ascii_lowercase + string.digits
+    return ''.join(random.choices(chars, k=length))
+
+
+def _generate_statsig_id() -> str:
+    """生成x-statsig-id
+    
+    随机选择两种格式：
+    1. e:TypeError: Cannot read properties of null (reading 'children['xxxxx']')
+    2. e:TypeError: Cannot read properties of undefined (reading 'xxxxxxxxxx')
+    """
+    if random.choice([True, False]):
+        rand = _random_string(5, letters_only=False)
+        msg = f"e:TypeError: Cannot read properties of null (reading 'children['{rand}']')"
+    else:
+        rand = _random_string(10)
+        msg = f"e:TypeError: Cannot read properties of undefined (reading '{rand}')"
+    
+    return base64.b64encode(msg.encode()).decode()
+
+
 def get_dynamic_headers(pathname: str = "/rest/app-chat/conversations/new") -> Dict[str, str]:
     """获取请求头
-
+    
     Args:
         pathname: 请求路径
-
+        
     Returns:
-        请求头字典
+        完整的请求头字典
     """
-    # 获取配置的 x-statsig-id
-    statsig_id = setting.grok_config.get("x_statsig_id")
-    if not statsig_id:
-        raise ValueError("配置文件中未设置 x_statsig_id")
+    # 获取或生成statsig-id
+    if setting.grok_config.get("dynamic_statsig", False):
+        statsig_id = _generate_statsig_id()
+        logger.debug(f"[Statsig] 动态生成: {statsig_id}")
+    else:
+        statsig_id = setting.grok_config.get("x_statsig_id")
+        if not statsig_id:
+            raise ValueError("配置文件中未设置 x_statsig_id")
+        logger.debug(f"[Statsig] 使用固定值: {statsig_id}")
 
-    # 构建基础请求头
-    headers = {
-        "Accept": "*/*",
-        "Accept-Language": "zh-CN,zh;q=0.9",
-        "Accept-Encoding": "gzip, deflate, br, zstd",
-        "Content-Type": "application/json" if "upload-file" not in pathname else "text/plain;charset=UTF-8",
-        "Connection": "keep-alive",
-        "Origin": "https://grok.com",
-        "Priority": "u=1, i",
-        "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/133.0.0.0 Safari/537.36",
-        "Sec-Ch-Ua": '"Not(A:Brand";v="99", "Google Chrome";v="133", "Chromium";v="133"',
-        "Sec-Ch-Ua-Mobile": "?0",
-        "Sec-Ch-Ua-Platform": '"macOS"',
-        "Sec-Fetch-Dest": "empty",
-        "Sec-Fetch-Mode": "cors",
-        "Sec-Fetch-Site": "same-origin",
-        "Baggage": "sentry-public_key=b311e0f2690c81f25e2c4cf6d4f7ce1c",
-        "x-statsig-id": statsig_id,
-        "x-xai-request-id": str(uuid.uuid4())
-    }
+    # 构建请求头
+    headers = BASE_HEADERS.copy()
+    headers["x-statsig-id"] = statsig_id
+    headers["x-xai-request-id"] = str(uuid.uuid4())
+    headers["Content-Type"] = "text/plain;charset=UTF-8" if "upload-file" in pathname else "application/json"
 
     return headers
\ No newline at end of file
diff --git a/app/services/grok/token.py b/app/services/grok/token.py
index a8678b0..80f2e98 100644
--- a/app/services/grok/token.py
+++ b/app/services/grok/token.py
@@ -1,9 +1,10 @@
-"""Grok Token 管理器模块"""
+"""Grok Token 管理器 - 单例模式的Token负载均衡和状态管理"""
 
-import json
+import orjson
 import time
 import asyncio
 import aiofiles
+import portalocker
 from pathlib import Path
 from curl_cffi.requests import AsyncSession
 from typing import Dict, Any, Optional, Tuple
@@ -14,37 +15,28 @@ from app.core.logger import logger
 from app.core.config import setting
 from app.services.grok.statsig import get_dynamic_headers
 
-# 常量定义
-RATE_LIMIT_ENDPOINT = "https://grok.com/rest/rate-limits"
-REQUEST_TIMEOUT = 30
-IMPERSONATE_BROWSER = "chrome133a"
-MAX_FAILURE_COUNT = 3
-TOKEN_INVALID_CODE = 401  # SSO Token失效
-STATSIG_INVALID_CODE = 403  # x-statsig-id失效
+
+# 常量
+RATE_LIMIT_API = "https://grok.com/rest/rate-limits"
+TIMEOUT = 30
+BROWSER = "chrome133a"
+MAX_FAILURES = 3
+TOKEN_INVALID = 401
+STATSIG_INVALID = 403
 
 
 class GrokTokenManager:
-    """
-    Grok Token管理器
-    
-    单例模式的Token管理器，负责：
-    - Token文件的读写操作
-    - Token负载均衡
-    - Token状态管理
-    - 支持普通Token和Super Token
-    """
+    """Token管理器（单例）"""
     
     _instance: Optional['GrokTokenManager'] = None
     _lock = asyncio.Lock()
 
     def __new__(cls) -> 'GrokTokenManager':
-        """单例模式实现"""
         if cls._instance is None:
             cls._instance = super().__new__(cls)
         return cls._instance
 
     def __init__(self):
-        """初始化Token管理器"""
         if hasattr(self, '_initialized'):
             return
 
@@ -52,79 +44,129 @@ class GrokTokenManager:
         self._file_lock = asyncio.Lock()
         self.token_file.parent.mkdir(parents=True, exist_ok=True)
         self._storage = None
-
-        # 同步加载初始数据
-        self._load_data()
+        self.token_data = None  # 延迟加载
+        
+        # 批量保存队列
+        self._save_pending = False  # 标记是否有待保存的数据
+        self._save_task = None  # 后台保存任务
+        self._shutdown = False  # 关闭标志
+        
         self._initialized = True
-
-        logger.debug(f"[Token] 管理器初始化完成，文件: {self.token_file}")
+        logger.debug(f"[Token] 初始化完成: {self.token_file}")
 
     def set_storage(self, storage) -> None:
         """设置存储实例"""
         self._storage = storage
 
-    def _load_data(self) -> None:
-        """同步加载Token数据（仅用于初始化）"""
-        default_data = {
-            TokenType.NORMAL.value: {},
-            TokenType.SUPER.value: {}
-        }
-
+    async def _load_data(self) -> None:
+        """异步加载Token数据（支持多进程）"""
+        default = {TokenType.NORMAL.value: {}, TokenType.SUPER.value: {}}
+        
         try:
             if self.token_file.exists():
-                with open(self.token_file, "r", encoding="utf-8") as f:
-                    self.token_data = json.load(f)
+                # 使用进程锁读取文件
+                async with self._file_lock:
+                    with open(self.token_file, "r", encoding="utf-8") as f:
+                        portalocker.lock(f, portalocker.LOCK_SH)  # 共享锁（读）
+                        try:
+                            content = f.read()
+                            self.token_data = orjson.loads(content)
+                        finally:
+                            portalocker.unlock(f)
             else:
-                self.token_data = default_data
-                logger.debug("[Token] 创建新的Token数据文件")
-        except (json.JSONDecodeError, IOError) as e:
-            logger.error(f"[Token] 加载Token数据失败: {str(e)}")
-            self.token_data = default_data
+                self.token_data = default
+                logger.debug("[Token] 创建新数据文件")
+        except Exception as e:
+            logger.error(f"[Token] 加载失败: {e}")
+            self.token_data = default
 
     async def _save_data(self) -> None:
-        """异步保存Token数据到存储"""
+        """保存Token数据（支持多进程）"""
         try:
             if not self._storage:
-                # 如果没有设置存储，使用传统文件保存方式（向后兼容）
                 async with self._file_lock:
-                    async with aiofiles.open(self.token_file, "w", encoding="utf-8") as f:
-                        await f.write(json.dumps(self.token_data, indent=2, ensure_ascii=False))
+                    # 使用进程锁写入文件
+                    with open(self.token_file, "w", encoding="utf-8") as f:
+                        portalocker.lock(f, portalocker.LOCK_EX)  # 独占锁（写）
+                        try:
+                            content = orjson.dumps(self.token_data, option=orjson.OPT_INDENT_2).decode()
+                            f.write(content)
+                            f.flush()  # 确保写入磁盘
+                        finally:
+                            portalocker.unlock(f)
             else:
-                # 使用存储抽象层
                 await self._storage.save_tokens(self.token_data)
         except IOError as e:
-            logger.error(f"[Token] 保存Token数据失败: {str(e)}")
-            raise GrokApiException(
-                f"Token数据保存失败: {str(e)}",
-                "TOKEN_SAVE_ERROR",
-                {"file_path": str(self.token_file)}
-            )
+            logger.error(f"[Token] 保存失败: {e}")
+            raise GrokApiException(f"保存失败: {e}", "TOKEN_SAVE_ERROR", {"file": str(self.token_file)})
+
+    def _mark_dirty(self) -> None:
+        """标记有待保存的数据"""
+        self._save_pending = True
+
+    async def _batch_save_worker(self) -> None:
+        """批量保存后台任务"""
+        from app.core.config import setting
+        
+        interval = setting.global_config.get("batch_save_interval", 1.0)
+        logger.info(f"[Token] 批量保存任务已启动，间隔: {interval}s")
+        
+        while not self._shutdown:
+            await asyncio.sleep(interval)
+            
+            if self._save_pending and not self._shutdown:
+                try:
+                    await self._save_data()
+                    self._save_pending = False
+                    logger.debug("[Token] 批量保存完成")
+                except Exception as e:
+                    logger.error(f"[Token] 批量保存失败: {e}")
+
+    async def start_batch_save(self) -> None:
+        """启动批量保存任务"""
+        if self._save_task is None:
+            self._save_task = asyncio.create_task(self._batch_save_worker())
+            logger.info("[Token] 批量保存任务已创建")
+
+    async def shutdown(self) -> None:
+        """关闭并刷新所有待保存数据"""
+        self._shutdown = True
+        
+        if self._save_task:
+            self._save_task.cancel()
+            try:
+                await self._save_task
+            except asyncio.CancelledError:
+                pass
+        
+        # 最终刷新
+        if self._save_pending:
+            await self._save_data()
+            logger.info("[Token] 关闭时刷新完成")
 
     @staticmethod
     def _extract_sso(auth_token: str) -> Optional[str]:
-        """从认证令牌中提取SSO值"""
+        """提取SSO值"""
         if "sso=" in auth_token:
             return auth_token.split("sso=")[1].split(";")[0]
-        logger.warning("[Token] 无法从认证令牌中提取SSO值")
+        logger.warning("[Token] 无法提取SSO值")
         return None
 
-    def _find_token(self, sso_value: str) -> Tuple[Optional[str], Optional[Dict[str, Any]]]:
-        """查找Token数据，返回(token_type, token_data)"""
+    def _find_token(self, sso: str) -> Tuple[Optional[str], Optional[Dict]]:
+        """查找Token"""
         for token_type in [TokenType.NORMAL.value, TokenType.SUPER.value]:
-            if sso_value in self.token_data[token_type]:
-                return token_type, self.token_data[token_type][sso_value]
+            if sso in self.token_data[token_type]:
+                return token_type, self.token_data[token_type][sso]
         return None, None
 
     async def add_token(self, tokens: list[str], token_type: TokenType) -> None:
-        """添加Token到管理器"""
+        """添加Token"""
         if not tokens:
-            logger.debug("[Token] 尝试添加空的Token列表")
             return
 
-        added_count = 0
+        count = 0
         for token in tokens:
             if not token or not token.strip():
-                logger.debug("[Token] 跳过空的Token")
                 continue
 
             self.token_data[token_type.value][token] = {
@@ -134,265 +176,312 @@ class GrokTokenManager:
                 "status": "active",
                 "failedCount": 0,
                 "lastFailureTime": None,
-                "lastFailureReason": None
+                "lastFailureReason": None,
+                "tags": [],
+                "note": ""
             }
-            added_count += 1
+            count += 1
 
-        await self._save_data()
-        logger.info(f"[Token] 成功添加 {added_count} 个 {token_type.value} Token")
+        self._mark_dirty()  # 批量保存
+        logger.info(f"[Token] 添加 {count} 个 {token_type.value} Token")
 
     async def delete_token(self, tokens: list[str], token_type: TokenType) -> None:
-        """删除指定的Token"""
+        """删除Token"""
         if not tokens:
-            logger.debug("[Token] 尝试删除空的Token列表")
             return
 
-        deleted_count = 0
+        count = 0
         for token in tokens:
             if token in self.token_data[token_type.value]:
                 del self.token_data[token_type.value][token]
-                deleted_count += 1
-            else:
-                logger.debug(f"[Token] Token不存在: {token[:10]}...")
+                count += 1
 
-        await self._save_data()
-        logger.info(f"[Token] 成功删除 {deleted_count} 个 {token_type.value} Token")
+        self._mark_dirty()  # 批量保存
+        logger.info(f"[Token] 删除 {count} 个 {token_type.value} Token")
+
+    async def update_token_tags(self, token: str, token_type: TokenType, tags: list[str]) -> None:
+        """更新Token标签"""
+        if token not in self.token_data[token_type.value]:
+            raise GrokApiException("Token不存在", "TOKEN_NOT_FOUND", {"token": token[:10]})
+        
+        cleaned = [t.strip() for t in tags if t and t.strip()]
+        self.token_data[token_type.value][token]["tags"] = cleaned
+        self._mark_dirty()  # 批量保存
+        logger.info(f"[Token] 更新标签: {token[:10]}... -> {cleaned}")
+
+    async def update_token_note(self, token: str, token_type: TokenType, note: str) -> None:
+        """更新Token备注"""
+        if token not in self.token_data[token_type.value]:
+            raise GrokApiException("Token不存在", "TOKEN_NOT_FOUND", {"token": token[:10]})
+        
+        self.token_data[token_type.value][token]["note"] = note.strip()
+        self._mark_dirty()  # 批量保存
+        logger.info(f"[Token] 更新备注: {token[:10]}...")
     
     def get_tokens(self) -> Dict[str, Any]:
-        """获取所有Token数据"""
+        """获取所有Token"""
         return self.token_data.copy()
 
+    def _reload_if_needed(self) -> None:
+        """在多进程模式下重新加载数据（同步版本，用于select_token）"""
+        # 只在文件模式且多进程环境下才重新加载
+        if self._storage:
+            return  # 数据库模式不需要
+        
+        try:
+            if self.token_file.exists():
+                with open(self.token_file, "r", encoding="utf-8") as f:
+                    portalocker.lock(f, portalocker.LOCK_SH)
+                    try:
+                        content = f.read()
+                        self.token_data = orjson.loads(content)
+                    finally:
+                        portalocker.unlock(f)
+        except Exception as e:
+            logger.warning(f"[Token] 重新加载失败: {e}")
+
     def get_token(self, model: str) -> str:
-        """获取指定模型的Token"""
-        jwt_token = self.select_token(model)
-        return f"sso-rw={jwt_token};sso={jwt_token}"
+        """获取Token"""
+        jwt = self.select_token(model)
+        return f"sso-rw={jwt};sso={jwt}"
     
     def select_token(self, model: str) -> str:
-        """根据模型类型和剩余次数选择最优Token"""
-        def select_best_token(tokens_dict: Dict[str, Any]) -> Tuple[Optional[str], Optional[int]]:
-            """从 token 字典中选择最佳 token"""
-            unused_tokens = []  # remaining = -1 的 token
-            used_tokens = []    # remaining > 0 的 token
+        """选择最优Token（多进程安全）"""
+        # 重新加载最新数据（多进程模式）
+        self._reload_if_needed()
+        def select_best(tokens: Dict[str, Any], field: str) -> Tuple[Optional[str], Optional[int]]:
+            """选择最佳Token"""
+            unused, used = [], []
 
-            for token_key, token_data in tokens_dict.items():
-                # 跳过已失效的Token
-                if token_data.get("status") == "expired":
+            for key, data in tokens.items():
+                # 跳过已失效的token
+                if data.get("status") == "expired":
+                    continue
+                
+                # 跳过失败次数过多的token（任何错误状态码）
+                if data.get("failedCount", 0) >= MAX_FAILURES:
                     continue
 
-                remaining = int(token_data.get(remaining_field, -1))
-
-                # 跳过已限流的 token
+                remaining = int(data.get(field, -1))
                 if remaining == 0:
                     continue
 
-                # 分类存储
                 if remaining == -1:
-                    unused_tokens.append(token_key)
+                    unused.append(key)
                 elif remaining > 0:
-                    used_tokens.append((token_key, remaining))
-
-            # 优先返回尚未使用的 token
-            if unused_tokens:
-                return unused_tokens[0], -1
-
-            # 否则返回次数最多的 token
-            if used_tokens:
-                used_tokens.sort(key=lambda x: x[1], reverse=True)
-                return used_tokens[0][0], used_tokens[0][1]
+                    used.append((key, remaining))
 
+            if unused:
+                return unused[0], -1
+            if used:
+                used.sort(key=lambda x: x[1], reverse=True)
+                return used[0][0], used[0][1]
             return None, None
 
-        max_token_key = None
-        max_remaining = None
-
-        # 深拷贝
-        token_data_snapshot = {
+        # 快照
+        snapshot = {
             TokenType.NORMAL.value: self.token_data[TokenType.NORMAL.value].copy(),
             TokenType.SUPER.value: self.token_data[TokenType.SUPER.value].copy()
         }
 
+        # 选择策略
         if model == "grok-4-heavy":
-            # grok-4-heavy 只能使用Super Token + heavy remaining queries
-            remaining_field = "heavyremainingQueries"
-            max_token_key, max_remaining = select_best_token(token_data_snapshot[TokenType.SUPER.value])
+            field = "heavyremainingQueries"
+            token_key, remaining = select_best(snapshot[TokenType.SUPER.value], field)
         else:
-            # 其他模型使用 remaining Queries
-            remaining_field = "remainingQueries"
+            field = "remainingQueries"
+            token_key, remaining = select_best(snapshot[TokenType.NORMAL.value], field)
+            if token_key is None:
+                token_key, remaining = select_best(snapshot[TokenType.SUPER.value], field)
 
-            # 优先使用普通Token
-            max_token_key, max_remaining = select_best_token(token_data_snapshot[TokenType.NORMAL.value])
-
-            # 如果普通Token没有可用的，尝试使用Super Token
-            if max_token_key is None:
-                max_token_key, max_remaining = select_best_token(token_data_snapshot[TokenType.SUPER.value])
-
-        if max_token_key is None:
+        if token_key is None:
             raise GrokApiException(
-                f"没有可用Token用于模型 {model}",
+                f"没有可用Token: {model}",
                 "NO_AVAILABLE_TOKEN",
                 {
                     "model": model,
-                    "normal_count": len(token_data_snapshot[TokenType.NORMAL.value]),
-                    "super_count": len(token_data_snapshot[TokenType.SUPER.value])
+                    "normal": len(snapshot[TokenType.NORMAL.value]),
+                    "super": len(snapshot[TokenType.SUPER.value])
                 }
             )
 
-        status_text = "未使用" if max_remaining == -1 else f"剩余{max_remaining}次"
-        logger.debug(f"[Token] 为模型 {model} 选择Token ({status_text})")
-        return max_token_key
+        status = "未使用" if remaining == -1 else f"剩余{remaining}次"
+        logger.debug(f"[Token] 分配Token: {model} ({status})")
+        return token_key
     
     async def check_limits(self, auth_token: str, model: str) -> Optional[Dict[str, Any]]:
-        """检查并更新模型速率限制"""
+        """检查速率限制"""
         try:
-            rate_limit_model_name = Models.to_rate_limit(model)
-            logger.debug(f"[Token] 检查模型 {model} (接口模型: {rate_limit_model_name}) 的速率限制")
-
-            # 准备请求
-            payload = {"requestKind": "DEFAULT", "modelName": rate_limit_model_name}
-            cf_clearance = setting.grok_config.get("cf_clearance", "")
-            cookie = f"{auth_token};{cf_clearance}" if cf_clearance else auth_token
-
+            rate_model = Models.to_rate_limit(model)
+            payload = {"requestKind": "DEFAULT", "modelName": rate_model}
+            
+            cf = setting.grok_config.get("cf_clearance", "")
             headers = get_dynamic_headers("/rest/rate-limits")
-            headers["Cookie"] = cookie
+            headers["Cookie"] = f"{auth_token};{cf}" if cf else auth_token
 
-            # 获取代理配置
-            proxy_url = setting.grok_config.get("proxy_url", "")
-            proxies = {"http": proxy_url, "https": proxy_url} if proxy_url else None
+            # 外层重试：可配置状态码（401/429等）
+            retry_codes = setting.grok_config.get("retry_status_codes", [401, 429])
+            MAX_OUTER_RETRY = 3
+            
+            for outer_retry in range(MAX_OUTER_RETRY + 1):  # +1 确保实际重试3次
+                # 内层重试：403代理池重试
+                max_403_retries = 5
+                retry_403_count = 0
+                
+                while retry_403_count <= max_403_retries:
+                    # 异步获取代理（支持代理池）
+                    from app.core.proxy_pool import proxy_pool
+                    
+                    # 如果是403重试且使用代理池，强制刷新代理
+                    if retry_403_count > 0 and proxy_pool._enabled:
+                        logger.info(f"[Token] 403重试 {retry_403_count}/{max_403_retries}，刷新代理...")
+                        proxy = await proxy_pool.force_refresh()
+                    else:
+                        proxy = await setting.get_proxy_async("service")
+                    
+                    proxies = {"http": proxy, "https": proxy} if proxy else None
+                    
+                    async with AsyncSession() as session:
+                        response = await session.post(
+                            RATE_LIMIT_API,
+                            headers=headers,
+                            json=payload,
+                            impersonate=BROWSER,
+                            timeout=TIMEOUT,
+                            proxies=proxies
+                        )
 
-            # 发送异步请求
-            async with AsyncSession() as session:
-                response = await session.post(
-                    RATE_LIMIT_ENDPOINT,
-                    headers=headers,
-                    json=payload,
-                    impersonate=IMPERSONATE_BROWSER,
-                    timeout=REQUEST_TIMEOUT,
-                    proxies=proxies
-                )
+                        # 内层403重试：仅当有代理池时触发
+                        if response.status_code == 403 and proxy_pool._enabled:
+                            retry_403_count += 1
+                            
+                            if retry_403_count <= max_403_retries:
+                                logger.warning(f"[Token] 遇到403错误，正在重试 ({retry_403_count}/{max_403_retries})...")
+                                await asyncio.sleep(0.5)
+                                continue
+                            
+                            # 内层重试全部失败
+                            logger.error(f"[Token] 403错误，已重试{retry_403_count-1}次，放弃")
+                            sso = self._extract_sso(auth_token)
+                            if sso:
+                                await self.record_failure(auth_token, 403, "服务器被Block")
+                        
+                        # 检查可配置状态码错误 - 外层重试
+                        if response.status_code in retry_codes:
+                            if outer_retry < MAX_OUTER_RETRY:
+                                delay = (outer_retry + 1) * 0.1  # 渐进延迟：0.1s, 0.2s, 0.3s
+                                logger.warning(f"[Token] 遇到{response.status_code}错误，外层重试 ({outer_retry+1}/{MAX_OUTER_RETRY})，等待{delay}s...")
+                                await asyncio.sleep(delay)
+                                break  # 跳出内层循环，进入外层重试
+                            else:
+                                logger.error(f"[Token] {response.status_code}错误，已重试{outer_retry}次，放弃")
+                                sso = self._extract_sso(auth_token)
+                                if sso:
+                                    if response.status_code == 401:
+                                        await self.record_failure(auth_token, 401, "Token失效")
+                                    else:
+                                        await self.record_failure(auth_token, response.status_code, f"错误: {response.status_code}")
+                                return None
 
-                if response.status_code == 200:
-                    rate_limit_data = response.json()
-                    logger.debug(f"[Token] 成功获取速率限制信息")
-
-                    # 保存速率限制信息
-                    sso_value = self._extract_sso(auth_token)
-                    if sso_value:
-                        if model == "grok-4-heavy":
-                            await self.update_limits(sso_value, normal=None, heavy=rate_limit_data.get("remainingQueries", -1))
-                            logger.info(f"[Token] 已更新限制: sso={sso_value[:10]}..., heavy={rate_limit_data.get('remainingQueries', -1)}")
+                        if response.status_code == 200:
+                            data = response.json()
+                            sso = self._extract_sso(auth_token)
+                            
+                            if outer_retry > 0 or retry_403_count > 0:
+                                logger.info(f"[Token] 重试成功！")
+                            
+                            if sso:
+                                if model == "grok-4-heavy":
+                                    await self.update_limits(sso, normal=None, heavy=data.get("remainingQueries", -1))
+                                    logger.info(f"[Token] 更新限制: {sso[:10]}..., heavy={data.get('remainingQueries', -1)}")
+                                else:
+                                    await self.update_limits(sso, normal=data.get("remainingTokens", -1), heavy=None)
+                                    logger.info(f"[Token] 更新限制: {sso[:10]}..., basic={data.get('remainingTokens', -1)}")
+                            
+                            return data
                         else:
-                            await self.update_limits(sso_value, normal=rate_limit_data.get("remainingTokens", -1), heavy=None)
-                            logger.info(f"[Token] 已更新限制: sso={sso_value[:10]}..., 通用={rate_limit_data.get('remainingTokens', -1)}")
-
-                    return rate_limit_data
-                else:
-                    logger.warning(f"[Token] 获取速率限制失败，状态码: {response.status_code}")
-                    return None
+                            # 其他错误
+                            logger.warning(f"[Token] 获取限制失败: {response.status_code}")
+                            sso = self._extract_sso(auth_token)
+                            if sso:
+                                await self.record_failure(auth_token, response.status_code, f"错误: {response.status_code}")
+                            return None
 
         except Exception as e:
-            logger.error(f"[Token] 检查速率限制时发生错误: {str(e)}")
+            logger.error(f"[Token] 检查限制错误: {e}")
             return None
 
-    async def update_limits(self, sso_value: str, normal: Optional[int] = None, heavy: Optional[int] = None) -> None:
-        """更新Token限制信息"""
+    async def update_limits(self, sso: str, normal: Optional[int] = None, heavy: Optional[int] = None) -> None:
+        """更新限制"""
         try:
             for token_type in [TokenType.NORMAL.value, TokenType.SUPER.value]:
-                if sso_value in self.token_data[token_type]:
+                if sso in self.token_data[token_type]:
                     if normal is not None:
-                        self.token_data[token_type][sso_value]["remainingQueries"] = normal
+                        self.token_data[token_type][sso]["remainingQueries"] = normal
                     if heavy is not None:
-                        self.token_data[token_type][sso_value]["heavyremainingQueries"] = heavy
-
-                    await self._save_data()
-                    logger.info(f"[Token] 已更新Token {sso_value[:10]}... 的限制信息")
+                        self.token_data[token_type][sso]["heavyremainingQueries"] = heavy
+                    self._mark_dirty()  # 批量保存
+                    logger.info(f"[Token] 更新限制: {sso[:10]}...")
                     return
-
-            logger.warning(f"[Token] 未找到SSO值为 {sso_value[:10]}... 的Token")
-
+            logger.warning(f"[Token] 未找到: {sso[:10]}...")
         except Exception as e:
-            logger.error(f"[Token] 更新Token限制时发生错误: {str(e)}")
+            logger.error(f"[Token] 更新限制错误: {e}")
     
-    async def record_failure(self, auth_token: str, status_code: int, error_message: str) -> None:
-        """记录Token失败信息
-
-        错误码说明：
-        - 401: SSO Token失效，会标记Token为expired
-        - 403: x-statsig-id失效，不影响Token状态
-
-        Args:
-            auth_token: 完整的认证Token (格式: sso-rw=xxx;sso=xxx)
-            status_code: HTTP状态码
-            error_message: 错误信息
-        """
+    async def record_failure(self, auth_token: str, status: int, msg: str) -> None:
+        """记录失败"""
         try:
-            # 403错误是x-statsig-id失效，不是Token问题
-            if status_code == STATSIG_INVALID_CODE:
-                logger.warning(f"[Token] x-statsig-id失效 (403)，需要更新配置文件中的x_statsig_id")
+            if status == STATSIG_INVALID:
+                logger.warning("[Token] IP被Block，请: 1.更换IP 2.使用代理 3.配置CF值")
                 return
 
-            sso_value = self._extract_sso(auth_token)
-            if not sso_value:
+            sso = self._extract_sso(auth_token)
+            if not sso:
                 return
 
-            _, token_data = self._find_token(sso_value)
-            if not token_data:
-                logger.warning(f"[Token] 未找到SSO值为 {sso_value[:10]}... 的Token")
+            _, data = self._find_token(sso)
+            if not data:
+                logger.warning(f"[Token] 未找到: {sso[:10]}...")
                 return
 
-            # 更新失败计数
-            token_data["failedCount"] = token_data.get("failedCount", 0) + 1
-            token_data["lastFailureTime"] = int(time.time() * 1000)
-            token_data["lastFailureReason"] = f"{status_code}: {error_message}"
+            data["failedCount"] = data.get("failedCount", 0) + 1
+            data["lastFailureTime"] = int(time.time() * 1000)
+            data["lastFailureReason"] = f"{status}: {msg}"
 
             logger.warning(
-                f"[Token] Token {sso_value[:10]}... 失败 (状态码: {status_code}), "
-                f"失败次数: {token_data['failedCount']}/{MAX_FAILURE_COUNT}, "
-                f"原因: {error_message}"
+                f"[Token] 失败: {sso[:10]}... (状态:{status}), "
+                f"次数: {data['failedCount']}/{MAX_FAILURES}, 原因: {msg}"
             )
 
-            # 只有401错误（SSO Token失效）且失败次数达到上限时，标记为失效
-            if status_code == TOKEN_INVALID_CODE and token_data["failedCount"] >= MAX_FAILURE_COUNT:
-                token_data["status"] = "expired"
-                logger.error(
-                    f"[Token] SSO Token {sso_value[:10]}... 已被标记为失效 "
-                    f"(连续401错误{token_data['failedCount']}次)"
-                )
+            if 400 <= status < 500 and data["failedCount"] >= MAX_FAILURES:
+                data["status"] = "expired"
+                logger.error(f"[Token] 标记失效: {sso[:10]}... (连续{status}错误{data['failedCount']}次)")
 
-            await self._save_data()
+            self._mark_dirty()  # 批量保存
 
         except Exception as e:
-            logger.error(f"[Token] 记录Token失败信息时发生错误: {str(e)}")
+            logger.error(f"[Token] 记录失败错误: {e}")
 
     async def reset_failure(self, auth_token: str) -> None:
-        """重置Token失败计数
-
-        当Token成功完成请求时调用此方法，用于清除失败记录。
-
-        Args:
-            auth_token: 完整的认证Token (格式: sso-rw=xxx;sso=xxx)
-        """
+        """重置失败计数"""
         try:
-            sso_value = self._extract_sso(auth_token)
-            if not sso_value:
+            sso = self._extract_sso(auth_token)
+            if not sso:
                 return
 
-            _, token_data = self._find_token(sso_value)
-            if not token_data:
-                logger.warning(f"[Token] 未找到SSO值为 {sso_value[:10]}... 的Token")
+            _, data = self._find_token(sso)
+            if not data:
                 return
 
-            # 只有在有失败记录时才重置并保存
-            if token_data.get("failedCount", 0) > 0:
-                token_data["failedCount"] = 0
-                token_data["lastFailureTime"] = None
-                token_data["lastFailureReason"] = None
-
-                await self._save_data()
-                logger.info(f"[Token] Token {sso_value[:10]}... 失败计数已重置")
+            if data.get("failedCount", 0) > 0:
+                data["failedCount"] = 0
+                data["lastFailureTime"] = None
+                data["lastFailureReason"] = None
+                self._mark_dirty()  # 批量保存
+                logger.info(f"[Token] 重置失败计数: {sso[:10]}...")
 
         except Exception as e:
-            logger.error(f"[Token] 重置Token失败计数时发生错误: {str(e)}")
+            logger.error(f"[Token] 重置失败错误: {e}")
 
 
-# 全局Token管理器实例
+# 全局实例
 token_manager = GrokTokenManager()
diff --git a/app/services/grok/upload.py b/app/services/grok/upload.py
index f5f92b3..4d0c8b4 100644
--- a/app/services/grok/upload.py
+++ b/app/services/grok/upload.py
@@ -1,10 +1,10 @@
-"""图片上传管理器"""
+"""图片上传管理器 - 支持Base64和URL图片上传"""
 
+import asyncio
 import base64
 import re
 from typing import Tuple, Optional
 from urllib.parse import urlparse
-
 from curl_cffi.requests import AsyncSession
 
 from app.services.grok.statsig import get_dynamic_headers
@@ -12,134 +12,198 @@ from app.core.exception import GrokApiException
 from app.core.config import setting
 from app.core.logger import logger
 
-# 常量定义
-UPLOAD_ENDPOINT = "https://grok.com/rest/app-chat/upload-file"
-REQUEST_TIMEOUT = 30
-IMPERSONATE_BROWSER = "chrome133a"
-DEFAULT_MIME_TYPE = "image/jpeg"
-DEFAULT_EXTENSION = "jpg"
+
+# 常量
+UPLOAD_API = "https://grok.com/rest/app-chat/upload-file"
+TIMEOUT = 30
+BROWSER = "chrome133a"
+
+# MIME类型
+MIME_TYPES = {
+    '.jpg': 'image/jpeg', '.jpeg': 'image/jpeg', '.png': 'image/png',
+    '.gif': 'image/gif', '.webp': 'image/webp', '.bmp': 'image/bmp',
+}
+DEFAULT_MIME = "image/jpeg"
+DEFAULT_EXT = "jpg"
 
 
 class ImageUploadManager:
-    """
-    Grok图片上传管理器
-    
-    提供图片上传功能，支持：
-    - Base64格式图片上传
-    - URL图片下载并上传
-    - 多种图片格式支持
-    """
+    """图片上传管理器"""
 
     @staticmethod
-    async def upload(image_input: str, auth_token: str) -> str:
-        """上传图片到Grok，支持Base64或URL"""
+    async def upload(image_input: str, auth_token: str) -> Tuple[str, str]:
+        """上传图片（支持Base64或URL）
+        
+        Returns:
+            (file_id, file_uri) 元组
+        """
         try:
+            # 判断类型并处理
             if ImageUploadManager._is_url(image_input):
-                # 下载 URL 图片
-                image_buffer, mime_type = await ImageUploadManager._download(image_input)
-
-                # 获取图片信息
-                file_name, _ = ImageUploadManager._get_info("", mime_type)
-
+                buffer, mime = await ImageUploadManager._download(image_input)
+                filename, _ = ImageUploadManager._get_info("", mime)
             else:
-                # 处理 base64 数据
-                image_buffer = image_input.split(",")[1] if "data:image" in image_input else image_input
+                buffer = image_input.split(",")[1] if "data:image" in image_input else image_input
+                filename, mime = ImageUploadManager._get_info(image_input)
 
-                # 获取图片信息
-                file_name, mime_type = ImageUploadManager._get_info(image_input)
-
-            # 构建上传数据
-            upload_data = {
-                "fileName": file_name,
-                "fileMimeType": mime_type,
-                "content": image_buffer,
+            # 构建数据
+            data = {
+                "fileName": filename,
+                "fileMimeType": mime,
+                "content": buffer,
             }
 
-            # 获取认证令牌
+
             if not auth_token:
-                raise GrokApiException("认证令牌缺失或为空", "NO_AUTH_TOKEN")
+                raise GrokApiException("认证令牌缺失", "NO_AUTH_TOKEN")
 
-            cf_clearance = setting.grok_config.get("cf_clearance", "")
-            cookie = f"{auth_token};{cf_clearance}" if cf_clearance else auth_token
-            proxy_url = setting.grok_config.get("proxy_url", "")
-            proxies = {"http": proxy_url, "https": proxy_url} if proxy_url else None
+            # 外层重试：可配置状态码（401/429等）
+            retry_codes = setting.grok_config.get("retry_status_codes", [401, 429])
+            MAX_OUTER_RETRY = 3
+            
+            for outer_retry in range(MAX_OUTER_RETRY + 1):  # +1 确保实际重试3次
+                try:
+                    # 内层重试：403代理池重试
+                    max_403_retries = 5
+                    retry_403_count = 0
+                    
+                    while retry_403_count <= max_403_retries:
+                        # 请求配置
+                        cf = setting.grok_config.get("cf_clearance", "")
+                        headers = {
+                            **get_dynamic_headers("/rest/app-chat/upload-file"),
+                            "Cookie": f"{auth_token};{cf}" if cf else auth_token,
+                        }
+                        
+                        # 异步获取代理（支持代理池）
+                        from app.core.proxy_pool import proxy_pool
+                        
+                        # 如果是403重试且使用代理池，强制刷新代理
+                        if retry_403_count > 0 and proxy_pool._enabled:
+                            logger.info(f"[Upload] 403重试 {retry_403_count}/{max_403_retries}，刷新代理...")
+                            proxy = await proxy_pool.force_refresh()
+                        else:
+                            proxy = await setting.get_proxy_async("service")
+                        
+                        proxies = {"http": proxy, "https": proxy} if proxy else None
 
-            # 发送异步请求
-            async with AsyncSession() as session:
-                response = await session.post(
-                    UPLOAD_ENDPOINT,
-                    headers={
-                        **get_dynamic_headers("/rest/app-chat/upload-file"),
-                        "Cookie": cookie,
-                    },
-                    json=upload_data,
-                    impersonate=IMPERSONATE_BROWSER,
-                    timeout=REQUEST_TIMEOUT,
-                    proxies=proxies,
-                )
+                        # 上传
+                        async with AsyncSession() as session:
+                            response = await session.post(
+                                UPLOAD_API,
+                                headers=headers,
+                                json=data,
+                                impersonate=BROWSER,
+                                timeout=TIMEOUT,
+                                proxies=proxies,
+                            )
 
-                # 检查响应
-                if response.status_code == 200:
-                    result = response.json()
-                    file_id = result.get("fileMetadataId", "")
-                    logger.debug(f"[Upload] 图片上传成功，文件ID: {file_id}")
-                    return file_id
-
-            return ""
+                            # 内层403重试：仅当有代理池时触发
+                            if response.status_code == 403 and proxy_pool._enabled:
+                                retry_403_count += 1
+                                
+                                if retry_403_count <= max_403_retries:
+                                    logger.warning(f"[Upload] 遇到403错误，正在重试 ({retry_403_count}/{max_403_retries})...")
+                                    await asyncio.sleep(0.5)
+                                    continue
+                                
+                                # 内层重试全部失败
+                                logger.error(f"[Upload] 403错误，已重试{retry_403_count-1}次，放弃")
+                            
+                            # 检查可配置状态码错误 - 外层重试
+                            if response.status_code in retry_codes:
+                                if outer_retry < MAX_OUTER_RETRY:
+                                    delay = (outer_retry + 1) * 0.1  # 渐进延迟：0.1s, 0.2s, 0.3s
+                                    logger.warning(f"[Upload] 遇到{response.status_code}错误，外层重试 ({outer_retry+1}/{MAX_OUTER_RETRY})，等待{delay}s...")
+                                    await asyncio.sleep(delay)
+                                    break  # 跳出内层循环，进入外层重试
+                                else:
+                                    logger.error(f"[Upload] {response.status_code}错误，已重试{outer_retry}次，放弃")
+                                    return "", ""
+                            
+                            if response.status_code == 200:
+                                result = response.json()
+                                file_id = result.get("fileMetadataId", "")
+                                file_uri = result.get("fileUri", "")
+                                
+                                if outer_retry > 0 or retry_403_count > 0:
+                                    logger.info(f"[Upload] 重试成功！")
+                                
+                                logger.debug(f"[Upload] 成功，ID: {file_id}")
+                                return file_id, file_uri
+                            
+                            # 其他错误直接返回
+                            logger.error(f"[Upload] 失败，状态码: {response._status_code}")
+                            return "", ""
+                    
+                    # 内层循环正常结束（非break），说明403重试全部失败
+                    return "", ""
+                
+                except Exception as e:
+                    if outer_retry < MAX_OUTER_RETRY - 1:
+                        logger.warning(f"[Upload] 异常: {e}，外层重试 ({outer_retry+1}/{MAX_OUTER_RETRY})...")
+                        await asyncio.sleep(0.5)
+                        continue
+                    
+                    logger.warning(f"[Upload] 失败: {e}")
+                    return "", ""
+            
+            return "", ""
 
         except Exception as e:
-            logger.warning(f"[Upload] 上传图片失败: {e}")
-            return ""
+            logger.warning(f"[Upload] 失败: {e}")
+            return "", ""
 
     @staticmethod
-    def _is_url(image_input: str) -> bool:
-        """检查输入是否为有效的URL"""
+    def _is_url(input_str: str) -> bool:
+        """检查是否为URL"""
         try:
-            result = urlparse(image_input)
+            result = urlparse(input_str)
             return all([result.scheme, result.netloc]) and result.scheme in ['http', 'https']
-        except Exception as e:
-            logger.warning(f"[Upload] URL解析失败: {e}")
+        except:
             return False
 
     @staticmethod
     async def _download(url: str) -> Tuple[str, str]:
-        """下载图片并转换为Base64"""
+        """下载图片并转Base64
+        
+        Returns:
+            (base64_string, mime_type) 元组
+        """
         try:
             async with AsyncSession() as session:
                 response = await session.get(url, timeout=5)
                 response.raise_for_status()
 
-                # 获取内容类型
-                content_type = response.headers.get('content-type', DEFAULT_MIME_TYPE)
+                content_type = response.headers.get('content-type', DEFAULT_MIME)
                 if not content_type.startswith('image/'):
-                    content_type = DEFAULT_MIME_TYPE
+                    content_type = DEFAULT_MIME
 
-                # 转换为 Base64
-                image_base64 = base64.b64encode(response.content).decode('utf-8')
-                return image_base64, content_type
+                b64 = base64.b64encode(response.content).decode()
+                return b64, content_type
         except Exception as e:
-            logger.warning(f"[Upload] 下载图片失败: {e}")
+            logger.warning(f"[Upload] 下载失败: {e}")
             return "", ""
 
     @staticmethod
     def _get_info(image_data: str, mime_type: Optional[str] = None) -> Tuple[str, str]:
-        """获取图片文件名和MIME类型"""
-        # mime_type 有值，直接使用
+        """获取文件名和MIME类型
+        
+        Returns:
+            (file_name, mime_type) 元组
+        """
+        # 已提供MIME类型
         if mime_type:
-            extension = mime_type.split("/")[1] if "/" in mime_type else DEFAULT_EXTENSION
-            file_name = f"image.{extension}"
-            return file_name, mime_type
+            ext = mime_type.split("/")[1] if "/" in mime_type else DEFAULT_EXT
+            return f"image.{ext}", mime_type
 
-        # mime_type 没有值，使用默认值
-        mime_type = DEFAULT_MIME_TYPE
-        extension = DEFAULT_EXTENSION
+        # 从Base64提取
+        mime = DEFAULT_MIME
+        ext = DEFAULT_EXT
 
-        # 从 Base64 数据中提取 MIME 类型
         if "data:image" in image_data:
-            match = re.search(r"data:([a-zA-Z0-9]+/[a-zA-Z0-9-.+]+);base64,", image_data)
-            if match:
-                mime_type = match.group(1)
-                extension = mime_type.split("/")[1]
+            if match := re.search(r"data:([a-zA-Z0-9]+/[a-zA-Z0-9-.+]+);base64,", image_data):
+                mime = match.group(1)
+                ext = mime.split("/")[1]
 
-        file_name = f"image.{extension}"
-        return file_name, mime_type
\ No newline at end of file
+        return f"image.{ext}", mime
\ No newline at end of file
diff --git a/app/services/mcp/__init__.py b/app/services/mcp/__init__.py
new file mode 100644
index 0000000..6951deb
--- /dev/null
+++ b/app/services/mcp/__init__.py
@@ -0,0 +1,6 @@
+# -*- coding: utf-8 -*-
+"""MCP模块初始化"""
+
+from app.services.mcp.server import mcp
+
+__all__ = ["mcp"]
diff --git a/app/services/mcp/server.py b/app/services/mcp/server.py
new file mode 100644
index 0000000..38d7719
--- /dev/null
+++ b/app/services/mcp/server.py
@@ -0,0 +1,63 @@
+# -*- coding: utf-8 -*-
+"""FastMCP服务器实例"""
+
+from fastmcp import FastMCP
+from fastmcp.server.auth.providers.jwt import StaticTokenVerifier
+from app.services.mcp.tools import ask_grok_impl
+from app.core.config import setting
+
+
+def create_mcp_server() -> FastMCP:
+    """创建MCP服务器实例，如果配置了API密钥则启用认证"""
+    # 检查是否配置了API密钥
+    api_key = setting.grok_config.get("api_key")
+    
+    # 如果配置了API密钥，则启用静态token验证
+    auth = None
+    if api_key:
+        auth = StaticTokenVerifier(
+            tokens={
+                api_key: {
+                    "client_id": "grok2api-client",
+                    "scopes": ["read", "write", "admin"]
+                }
+            },
+            required_scopes=["read"]
+        )
+    
+    # 创建FastMCP实例
+    return FastMCP(
+        name="Grok2API-MCP",
+        instructions="MCP server providing Grok AI chat capabilities. Use ask_grok tool to interact with Grok AI models.",
+        auth=auth
+    )
+
+
+# 创建全局MCP实例
+mcp = create_mcp_server()
+
+
+# 注册ask_grok工具
+@mcp.tool
+async def ask_grok(
+    query: str,
+    model: str = "grok-3-fast",
+    system_prompt: str = None
+) -> str:
+    """
+    调用Grok AI进行对话，尤其适用于当用户询问最新信息，需要调用搜索功能，或是想了解社交平台动态（如Twitter(X)、Reddit等）时。
+
+    Args:
+        query: 用户的问题或指令
+        model: Grok模型名称,可选值: grok-3-fast(默认), grok-4-fast, grok-4-fast-expert, grok-4-expert, grok-4-heavy
+        system_prompt: 可选的系统提示词,用于设定AI的角色或行为约束
+
+    Returns:
+        Grok AI的完整回复内容,可能包括文本和图片链接(Markdown格式)
+
+    Examples:
+        - 简单问答: ask_grok("什么是Python?")
+        - 指定模型: ask_grok("解释量子计算", model="grok-4-fast")
+        - 带系统提示: ask_grok("写一首诗", system_prompt="你是一位古典诗人")
+    """
+    return await ask_grok_impl(query, model, system_prompt)
diff --git a/app/services/mcp/tools.py b/app/services/mcp/tools.py
new file mode 100644
index 0000000..abedfae
--- /dev/null
+++ b/app/services/mcp/tools.py
@@ -0,0 +1,77 @@
+# -*- coding: utf-8 -*-
+"""MCP Tools - Grok AI 对话工具"""
+
+import json
+from typing import Optional
+from app.services.grok.client import GrokClient
+from app.core.logger import logger
+from app.core.exception import GrokApiException
+
+
+async def ask_grok_impl(
+    query: str,
+    model: str = "grok-3-fast",
+    system_prompt: Optional[str] = None
+) -> str:
+    """
+    内部实现: 调用Grok API并收集完整响应
+
+    Args:
+        query: 用户问题
+        model: 模型名称
+        system_prompt: 系统提示词
+
+    Returns:
+        str: 完整的Grok响应内容
+    """
+    try:
+        # 构建消息列表
+        messages = []
+        if system_prompt:
+            messages.append({"role": "system", "content": system_prompt})
+        messages.append({"role": "user", "content": query})
+
+        # 构建请求
+        request_data = {
+            "model": model,
+            "messages": messages,
+            "stream": True
+        }
+
+        logger.info(f"[MCP] ask_grok 调用, 模型: {model}")
+
+        # 调用Grok客户端(流式)
+        response_iterator = await GrokClient.openai_to_grok(request_data)
+
+        # 收集所有流式响应块
+        content_parts = []
+        async for chunk in response_iterator:
+            if isinstance(chunk, bytes):
+                chunk = chunk.decode('utf-8')
+
+            # 解析SSE格式
+            if chunk.startswith("data: "):
+                data_str = chunk[6:].strip()
+                if data_str == "[DONE]":
+                    break
+
+                try:
+                    data = json.loads(data_str)
+                    choices = data.get("choices", [])
+                    if choices:
+                        delta = choices[0].get("delta", {})
+                        if content := delta.get("content"):
+                            content_parts.append(content)
+                except json.JSONDecodeError:
+                    continue
+
+        result = "".join(content_parts)
+        logger.info(f"[MCP] ask_grok 完成, 响应长度: {len(result)}")
+        return result
+
+    except GrokApiException as e:
+        logger.error(f"[MCP] Grok API错误: {str(e)}")
+        raise Exception(f"Grok API调用失败: {str(e)}")
+    except Exception as e:
+        logger.error(f"[MCP] ask_grok异常: {str(e)}", exc_info=True)
+        raise Exception(f"处理请求时出错: {str(e)}")
diff --git a/app/template/admin.html b/app/template/admin.html
index 41e850d..0e47cc9 100644
--- a/app/template/admin.html
+++ b/app/template/admin.html
@@ -1,530 +1,887 @@
 <!DOCTYPE html>
 <html lang="zh-CN" class="h-full">
+
 <head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>管理控制台 - Grok2API</title>
-    <link rel="icon" type="image/png" href="/static/favicon.png">
-    <script src="https://cdn.tailwindcss.com"></script>
-    <style>
-        @keyframes slide-up{from{transform:translateY(100%);opacity:0}to{transform:translateY(0);opacity:1}}
-        .animate-slide-up{animation:slide-up .3s ease-out}
-        .tab-btn{transition:all .2s ease}
-        .hover-card{position:relative;display:inline-block}
-        .hover-card-trigger{cursor:pointer}
-        .hover-card-content{position:absolute;left:50%;transform:translateX(-50%);background:hsl(0 0% 3.9%);color:hsl(0 0% 98%);padding:8px 12px;border-radius:6px;font-size:12px;font-weight:500;white-space:nowrap;z-index:9999;pointer-events:none;opacity:0;visibility:hidden;transition:opacity .2s ease,transform .2s ease;box-shadow:0 4px 12px rgba(0,0,0,.25);border:1px solid hsl(0 0% 14.9%)}
-        .hover-card:hover .hover-card-content{opacity:1;visibility:visible}
-        .hover-card-content.top{bottom:100%;transform:translateX(-50%) translateY(-8px)}
-        .hover-card-content.bottom{top:100%;transform:translateX(-50%) translateY(8px)}
-        .hover-card-content::after{content:'';position:absolute;left:50%;transform:translateX(-50%);border:6px solid transparent;z-index:1000}
-        .hover-card-content.top::after{top:100%;border-top-color:hsl(0 0% 3.9%)}
-        .hover-card-content.bottom::after{bottom:100%;border-bottom-color:hsl(0 0% 3.9%)}
-        .hover-card-trigger:hover+.hover-card-content{opacity:1;visibility:visible}
-        [title]{position:relative}
-        [title]:hover::after{content:attr(title);position:absolute;bottom:100%;left:50%;transform:translateX(-50%) translateY(-4px);background:hsl(0 0% 3.9%);color:hsl(0 0% 98%);padding:4px 8px;border-radius:4px;font-size:11px;white-space:nowrap;z-index:1000;pointer-events:none;box-shadow:0 2px 8px rgba(0,0,0,.15)}
-        [title]:hover::before{content:'';position:absolute;bottom:100%;left:50%;transform:translateX(-50%);border:4px solid transparent;border-top-color:hsl(0 0% 3.9%);z-index:1000}
-    </style>
-    <script>
-        tailwind.config={theme:{extend:{colors:{border:"hsl(0 0% 89%)",input:"hsl(0 0% 89%)",ring:"hsl(0 0% 3.9%)",background:"hsl(0 0% 100%)",foreground:"hsl(0 0% 3.9%)",primary:{DEFAULT:"hsl(0 0% 9%)",foreground:"hsl(0 0% 98%)"},secondary:{DEFAULT:"hsl(0 0% 96.1%)",foreground:"hsl(0 0% 9%)"},muted:{DEFAULT:"hsl(0 0% 96.1%)",foreground:"hsl(0 0% 45.1%)"},accent:{DEFAULT:"hsl(0 0% 96.1%)",foreground:"hsl(0 0% 9%)"},destructive:{DEFAULT:"hsl(0 84.2% 60.2%)",foreground:"hsl(0 0% 98%)"}}}}}
-    </script>
+  <meta charset="UTF-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  <title>管理控制台 - Grok2API</title>
+  <link rel="icon" type="image/png" href="/static/favicon.png">
+  <script src="https://cdn.tailwindcss.com"></script>
+  <style>
+    @keyframes slide-up {
+      from {
+        transform: translateY(100%);
+        opacity: 0
+      }
+
+      to {
+        transform: translateY(0);
+        opacity: 1
+      }
+    }
+
+    .animate-slide-up {
+      animation: slide-up .3s ease-out
+    }
+
+    .tab-btn {
+      transition: all .2s ease
+    }
+
+    .hover-card {
+      position: relative;
+      display: inline-block
+    }
+
+    .hover-card-trigger {
+      cursor: pointer
+    }
+
+    .hover-card-content {
+      position: absolute;
+      left: 50%;
+      transform: translateX(-50%);
+      background: hsl(0 0% 3.9%);
+      color: hsl(0 0% 98%);
+      padding: 8px 12px;
+      border-radius: 6px;
+      font-size: 12px;
+      font-weight: 500;
+      white-space: nowrap;
+      z-index: 9999;
+      pointer-events: none;
+      opacity: 0;
+      visibility: hidden;
+      transition: opacity .2s ease, transform .2s ease;
+      box-shadow: 0 4px 12px rgba(0, 0, 0, .25);
+      border: 1px solid hsl(0 0% 14.9%)
+    }
+
+    .hover-card:hover .hover-card-content {
+      opacity: 1;
+      visibility: visible
+    }
+
+    .hover-card-content.top {
+      bottom: 100%;
+      transform: translateX(-50%) translateY(-8px)
+    }
+
+    .hover-card-content.bottom {
+      top: 100%;
+      transform: translateX(-50%) translateY(8px)
+    }
+
+    .hover-card-content::after {
+      content: '';
+      position: absolute;
+      left: 50%;
+      transform: translateX(-50%);
+      border: 6px solid transparent;
+      z-index: 1000
+    }
+
+    .hover-card-content.top::after {
+      top: 100%;
+      border-top-color: hsl(0 0% 3.9%)
+    }
+
+    .hover-card-content.bottom::after {
+      bottom: 100%;
+      border-bottom-color: hsl(0 0% 3.9%)
+    }
+
+    .hover-card-trigger:hover+.hover-card-content {
+      opacity: 1;
+      visibility: visible
+    }
+
+    .btn-icon {
+      display: inline-flex;
+      align-items: center;
+      justify-content: center;
+      border-radius: 0.375rem;
+      transition: color .2s, background-color .2s;
+      height: 2rem;
+      width: 2rem
+    }
+
+    .btn-icon:focus-visible {
+      outline: 2px solid transparent;
+      outline-offset: 2px;
+      box-shadow: 0 0 0 1px hsl(0 0% 3.9%)
+    }
+
+    .btn-icon:hover {
+      background-color: hsl(0 0% 96.1%);
+      color: hsl(0 0% 9%)
+    }
+
+    .sticky-right {
+      position: sticky;
+      right: 0;
+      background-color: hsl(0 0% 100%);
+      z-index: 10;
+      border-left: 1px solid hsl(0 0% 89%)
+    }
+
+    .cfg-label {
+      font-size: 0.875rem;
+      font-weight: 500;
+      color: hsl(0 0% 45.1%);
+      margin-bottom: 0.5rem;
+      display: flex;
+      align-items: center;
+      gap: 0.25rem
+    }
+
+    .cfg-input {
+      display: flex;
+      height: 2.25rem;
+      width: 100%;
+      border-radius: 0.375rem;
+      border: 1px solid hsl(0 0% 89%);
+      background-color: hsl(0 0% 100%);
+      padding: 0.5rem 0.75rem;
+      font-size: 0.875rem
+    }
+
+    .help-icon {
+      display: inline-flex;
+      align-items: center;
+      justify-content: center;
+      width: 0.875rem;
+      height: 0.875rem;
+      border-radius: 9999px;
+      border: 1px solid hsl(0 0% 45.1%);
+      color: hsl(0 0% 45.1%);
+      cursor: help;
+      font-size: 10px;
+      line-height: 1
+    }
+
+    [title] {
+      position: relative
+    }
+
+    [title]:hover::after {
+      content: attr(title);
+      position: absolute;
+      bottom: 100%;
+      left: 50%;
+      transform: translateX(-50%) translateY(-8px);
+      background: hsl(0 0% 11%);
+      color: hsl(0 0% 98%);
+      padding: 8px 12px;
+      border-radius: 6px;
+      font-size: 12px;
+      font-weight: 500;
+      line-height: 1.4;
+      white-space: pre-line;
+      max-width: 350px;
+      width: max-content;
+      word-wrap: break-word;
+      z-index: 1000;
+      pointer-events: none;
+      box-shadow: 0 8px 20px rgba(0, 0, 0, 0.15), 0 2px 8px rgba(0, 0, 0, 0.1);
+      opacity: 0;
+      visibility: hidden;
+      transition: all 0.2s ease;
+      animation: tooltipFadeIn 0.2s ease forwards
+    }
+
+    [title]:hover::before {
+      content: '';
+      position: absolute;
+      bottom: 100%;
+      left: 50%;
+      transform: translateX(-50%) translateY(-4px);
+      border: 6px solid transparent;
+      border-top-color: hsl(0 0% 11%);
+      z-index: 1000;
+      opacity: 0;
+      visibility: hidden;
+      transition: opacity 0.2s ease, visibility 0.2s ease
+    }
+
+    [title]:hover::after,
+    [title]:hover::before {
+      opacity: 1;
+      visibility: visible
+    }
+
+    @keyframes tooltipFadeIn {
+      from {
+        opacity: 0;
+        transform: translateX(-50%) translateY(-4px)
+      }
+
+      to {
+        opacity: 1;
+        transform: translateX(-50%) translateY(-8px)
+      }
+    }
+  </style>
+  <script>
+    tailwind.config = { theme: { extend: { colors: { border: "hsl(0 0% 89%)", input: "hsl(0 0% 89%)", ring: "hsl(0 0% 3.9%)", background: "hsl(0 0% 100%)", foreground: "hsl(0 0% 3.9%)", primary: { DEFAULT: "hsl(0 0% 9%)", foreground: "hsl(0 0% 98%)" }, secondary: { DEFAULT: "hsl(0 0% 96.1%)", foreground: "hsl(0 0% 9%)" }, muted: { DEFAULT: "hsl(0 0% 96.1%)", foreground: "hsl(0 0% 45.1%)" }, accent: { DEFAULT: "hsl(0 0% 96.1%)", foreground: "hsl(0 0% 9%)" }, destructive: { DEFAULT: "hsl(0 84.2% 60.2%)", foreground: "hsl(0 0% 98%)" } } } } }
+  </script>
 </head>
+
 <body class="h-full bg-background text-foreground antialiased">
-    <!-- 导航栏 -->
-    <header class="sticky top-0 z-50 w-full border-b border-border/40 bg-background/95 backdrop-blur supports-[backdrop-filter]:bg-background/60">
-        <div class="mx-auto flex h-14 max-w-7xl items-center px-6">
-            <div class="mr-4 flex items-baseline gap-3">
-                <span class="font-bold text-xl">Grok2API</span>
-                <span class="text-xs text-gray-400">by @Chenyme</span>
-            </div>
-            <div class="flex flex-1 items-center justify-end">
-                <div class="flex items-center gap-2">
-                    <div class="hover-card">
-                        <span id="storageMode" class="hover-card-trigger inline-flex items-center rounded-full px-2 py-0.5 text-xs font-medium bg-muted text-muted-foreground border">
-                            <svg class="h-3 w-3 mr-1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                                <ellipse cx="12" cy="5" rx="9" ry="3"/>
-                                <path d="m21 12c0 1.66-4 3-9 3s-9-1.34-9-3"/>
-                                <path d="m3 5v14c0 1.66 4 3 9 3s9-1.34 9-3V5"/>
-                            </svg>
-                            <span id="storageModeText">FILE</span>
-                        </span>
-                        <div id="storageModeTooltip" class="hover-card-content top">
-                            加载中...
-                        </div>
-                    </div>
-                    <nav class="flex items-center gap-1">
-                    <a
-                        href="https://github.com/chenyme/grok2api/issues"
-                        target="_blank"
-                        class="inline-flex items-center justify-center text-xs transition-colors hover:bg-accent hover:text-accent-foreground h-7 px-2.5 gap-1"
-                    >
-                        <svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                            <path d="M21 15a2 2 0 0 1-2 2H7l-4 4V5a2 2 0 0 1 2-2h14a2 2 0 0 1 2 2z"/>
-                        </svg>
-                        反馈
-                    </a>
-                    <button
-                        onclick="logout()"
-                        class="inline-flex items-center justify-center text-xs transition-colors hover:bg-accent hover:text-accent-foreground h-7 px-2.5 gap-1"
-                    >
-                        <svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                            <path d="M9 21H5a2 2 0 0 1-2-2V5a2 2 0 0 1 2-2h4"/>
-                            <polyline points="16 17 21 12 16 7"/>
-                            <line x1="21" y1="12" x2="9" y2="12"/>
-                        </svg>
-                        退出
-                    </button>
-                </nav>
+  <!-- 导航栏 -->
+  <header
+    class="sticky top-0 z-50 w-full border-b border-border/40 bg-background/95 backdrop-blur supports-[backdrop-filter]:bg-background/60">
+    <div class="mx-auto flex h-14 max-w-7xl items-center px-6">
+      <div class="mr-4 flex items-baseline gap-3">
+        <span class="font-bold text-xl">Grok2API</span>
+        <span class="text-xs text-gray-400">by @Chenyme</span>
+      </div>
+      <div class="flex flex-1 items-center justify-end">
+        <div class="flex items-center gap-2">
+          <div class="hover-card">
+            <span id="storageMode"
+              class="hover-card-trigger inline-flex items-center rounded-full px-2 py-0.5 text-xs font-medium bg-muted text-muted-foreground border">
+              <svg class="h-3 w-3 mr-1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                stroke="currentColor" stroke-width="2">
+                <ellipse cx="12" cy="5" rx="9" ry="3" />
+                <path d="m21 12c0 1.66-4 3-9 3s-9-1.34-9-3" />
+                <path d="m3 5v14c0 1.66 4 3 9 3s9-1.34 9-3V5" />
+              </svg>
+              <span id="storageModeText">FILE</span>
+            </span>
+            <div id="storageModeTooltip" class="hover-card-content top">
+              加载中...
             </div>
+          </div>
+          <nav class="flex items-center gap-1">
+            <a href="https://github.com/chenyme/grok2api/issues" target="_blank"
+              class="inline-flex items-center justify-center text-xs transition-colors hover:bg-accent hover:text-accent-foreground h-7 px-2.5 gap-1">
+              <svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                stroke="currentColor" stroke-width="2">
+                <path d="M21 15a2 2 0 0 1-2 2H7l-4 4V5a2 2 0 0 1 2-2h14a2 2 0 0 1 2 2z" />
+              </svg>
+              反馈
+            </a>
+            <button onclick="logout()"
+              class="inline-flex items-center justify-center text-xs transition-colors hover:bg-accent hover:text-accent-foreground h-7 px-2.5 gap-1">
+              <svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                stroke="currentColor" stroke-width="2">
+                <path d="M9 21H5a2 2 0 0 1-2-2V5a2 2 0 0 1 2-2h4" />
+                <polyline points="16 17 21 12 16 7" />
+                <line x1="21" y1="12" x2="9" y2="12" />
+              </svg>
+              退出
+            </button>
+          </nav>
         </div>
-    </header>
+      </div>
+  </header>
 
-    <main class="mx-auto max-w-7xl px-6 py-6">
-        <!-- Tab 导航 -->
-        <div class="border-b border-border mb-6">
-            <nav class="flex space-x-8">
-                <button onclick="switchTab('tokens')" id="tabTokens" class="tab-btn border-b-2 border-primary text-sm font-medium py-3 px-1">Token 管理</button>
-                <button onclick="switchTab('settings')" id="tabSettings" class="tab-btn border-b-2 border-transparent text-sm font-medium py-3 px-1">Setting 配置</button>
-            </nav>
-        </div>
-
-        <!-- Token 管理面板 -->
-        <div id="panelTokens">
-        <!-- 统计卡片 -->
-        <div class="grid gap-4 grid-cols-2 md:grid-cols-4 mb-6">
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">Token 总数</p>
-                <h3 class="text-xl font-bold" id="statTotal">-</h3>
-            </div>
-
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">未使用</p>
-                <h3 class="text-xl font-bold text-gray-500" id="statUnused">-</h3>
-            </div>
-
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">限流中</p>
-                <h3 class="text-xl font-bold text-orange-600" id="statLimited">-</h3>
-            </div>
-
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">失效</p>
-                <h3 class="text-xl font-bold text-destructive" id="statExpired">-</h3>
-            </div>
-
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">正常</p>
-                <h3 class="text-xl font-bold text-green-600" id="statActive">-</h3>
-            </div>
-
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">总剩余</p>
-                <h3 class="text-xl font-bold" id="statTotalRemaining">-</h3>
-            </div>
-
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">普通剩余</p>
-                <h3 class="text-xl font-bold text-blue-600" id="statNormalRemaining">-</h3>
-            </div>
-
-            <div class="rounded-lg border border-border bg-background p-4">
-                <p class="text-sm font-medium text-muted-foreground mb-2">高级剩余</p>
-                <h3 class="text-xl font-bold text-purple-600" id="statHeavyRemaining">-</h3>
-            </div>
-        </div>
-
-        <!-- Token 列表 -->
-        <div class="rounded-lg border border-border bg-background">
-            <!-- 工具栏 -->
-            <div class="flex items-center justify-between gap-4 p-4 border-b border-border">
-                <div class="flex items-center gap-3 flex-1">
-                    <div class="flex items-center gap-2">
-                        <select id="filterType" onchange="filterTokens()" class="h-8 px-3 text-sm rounded-md border border-input bg-background focus:outline-none focus:ring-1 focus:ring-ring">
-                            <option value="all">全部类型</option>
-                            <option value="sso">SSO</option>
-                            <option value="ssoSuper">SuperSSO</option>
-                        </select>
-                        <select id="filterStatus" onchange="filterTokens()" class="h-8 px-3 text-sm rounded-md border border-input bg-background focus:outline-none focus:ring-1 focus:ring-ring">
-                            <option value="all">全部状态</option>
-                            <option value="未使用">未使用</option>
-                            <option value="限流中">限流中</option>
-                            <option value="失效">失效</option>
-                            <option value="正常">正常</option>
-                        </select>
-                    </div>
-                </div>
-
-                <div class="flex items-center gap-2">
-                    <button 
-                        onclick="refreshTokens()"
-                        class="inline-flex items-center justify-center rounded-md transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-accent hover:text-accent-foreground h-8 w-8"
-                        title="刷新列表"
-                    >
-                        <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                            <polyline points="23 4 23 10 17 10"/>
-                            <polyline points="1 20 1 14 7 14"/>
-                            <path d="M3.51 9a9 9 0 0 1 14.85-3.36L23 10M1 14l4.64 4.36A9 9 0 0 0 20.49 15"/>
-                        </svg>
-                    </button>
-                    <div id="batchActions" class="hidden items-center gap-2">
-                        <button
-                            onclick="exportSelected()"
-                            class="inline-flex items-center justify-center rounded-md transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-accent hover:text-accent-foreground h-8 w-8"
-                            title="导出选中项"
-                        >
-                            <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                                <path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/>
-                                <polyline points="7 10 12 15 17 10"/>
-                                <line x1="12" y1="15" x2="12" y2="3"/>
-                            </svg>
-                        </button>
-                        <button
-                            onclick="batchDelete()"
-                            class="inline-flex items-center justify-center rounded-md transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-destructive/10 hover:text-destructive h-8 w-8"
-                            title="批量删除"
-                        >
-                            <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                                <polyline points="3 6 5 6 21 6"/>
-                                <path d="M19 6v14a2 2 0 0 1-2 2H7a2 2 0 0 1-2-2V6m3 0V4a2 2 0 0 1 2-2h4a2 2 0 0 1 2 2v2"/>
-                                <line x1="10" y1="11" x2="10" y2="17"/>
-                                <line x1="14" y1="11" x2="14" y2="17"/>
-                            </svg>
-                        </button>
-                    </div>
-                    <button
-                        onclick="openAddModal()"
-                        class="inline-flex items-center justify-center rounded-md transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring bg-primary text-primary-foreground hover:bg-primary/90 h-8 px-3"
-                        title="添加 Token"
-                    >
-                        <svg class="h-4 w-4 mr-2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                            <line x1="12" y1="5" x2="12" y2="19"/>
-                            <line x1="5" y1="12" x2="19" y2="12"/>
-                        </svg>
-                        <span class="text-sm font-medium">新增</span>
-                    </button>
-                </div>
-            </div>
-            
-            <!-- 表格 -->
-            <div class="relative w-full overflow-auto">
-                <table class="w-full text-sm table-fixed">
-                    <thead>
-                        <tr class="border-b border-border">
-                            <th class="h-10 px-3 text-left align-middle font-medium w-12">
-                                <input
-                                    type="checkbox"
-                                    id="selectAll"
-                                    onchange="toggleSelectAll()"
-                                    class="h-3.5 w-3.5 rounded border border-input focus:ring-1 focus:ring-ring"
-                                >
-                            </th>
-                            <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-72">Token</th>
-                            <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">类型</th>
-                            <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">状态</th>
-                            <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">普通剩余</th>
-                            <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">高级剩余</th>
-                            <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-40">创建时间</th>
-                            <th class="h-10 px-3 text-right align-middle text-sm font-medium text-muted-foreground w-20">操作</th>
-                        </tr>
-                    </thead>
-                    <tbody id="tokenTableBody" class="divide-y divide-border">
-                        <!-- 动态填充 -->
-                    </tbody>
-                </table>
-            </div>
-            
-            <div id="emptyState" class="hidden flex flex-col items-center justify-center py-12">
-                <svg class="h-10 w-10 text-muted-foreground/50 mb-3" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                    <rect x="3" y="3" width="18" height="18" rx="2"/>
-                    <path d="M9 9h6v6H9z"/>
-                </svg>
-                <p class="text-sm text-muted-foreground">暂无数据</p>
-            </div>
-        </div>
-        </div>
-
-        <!-- 全局设置面板 -->
-        <div id="panelSettings" class="hidden">
-            <!-- 配置提示 -->
-            <div class="mb-6 rounded border border-blue-300 bg-blue-50 px-4 py-3">
-                <div class="text-xs text-gray-800 leading-relaxed">
-                    <div class="text-base font-medium text-gray-900 mb-2.5">部分说明</div>
-                    <div class="space-y-1.5">
-                        <div><span class="font-medium">服务网址：</span>图片/视频链接返回时需要拼接您的服务网址（如 https://yourdomain.com），若您不使用视频功能且图片使用Base64模式则可留空</div>
-                        <div><span class="font-medium">日志报错 403：</span>通常是被 CF 拦截了，可配置代理IP、更换服务器IP或者在服务器中访问 grok.com 通过 CF 验证后 F12 获取 cf_clearance</div>
-                        <div><span class="font-medium">X Statsig ID：</span>反机器人验证参数，勿随意修改，一般 403 是被盾了，并不是需要更新 X Statsig ID</div>
-                    </div>
-                </div>
-            </div>
-
-            <div class="grid gap-6 lg:grid-cols-2">
-                <!-- 全局配置 -->
-                <div class="rounded-lg border border-border bg-background p-6">
-                    <h3 class="text-lg font-semibold mb-4">全局配置</h3>
-                    <div class="space-y-4">
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                管理账户
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="登录管理后台的用户名">?</span>
-                            </label>
-                            <input id="cfgAdminUser" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="admin">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                管理密码
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="登录管理后台的密码，留空表示不修改当前密码">?</span>
-                            </label>
-                            <input id="cfgAdminPass" type="password" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="留空则不修改">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                日志级别
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="日志详细程度。DEBUG：最详细 | INFO：一般信息 | WARNING：警告 | ERROR：仅错误">?</span>
-                            </label>
-                            <select id="cfgLogLevel" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm">
-                                <option>DEBUG</option><option>INFO</option><option>WARNING</option><option>ERROR</option>
-                            </select>
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                图片模式
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="选择图片返回方式：URL模式返回图片链接（需缓存）；Base64模式返回base64编码（无需缓存，直接嵌入响应）">?</span>
-                            </label>
-                            <select id="cfgImageMode" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring">
-                                <option value="url">URL (图片链接)</option>
-                                <option value="base64">Base64 (base64编码)</option>
-                            </select>
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                服务网址
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="用于生成图片链接的服务地址。建议使用域名以保护服务器IP，如无图片需求可留空。注意：Base64模式下无需填写此项">?</span>
-                            </label>
-                            <input id="cfgBaseUrl" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="http://localhost:8000">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                图片缓存上限 (MB)
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="图片缓存目录的最大容量，超过后自动删除最旧的缓存文件">?</span>
-                            </label>
-                            <input id="cfgImageCacheMaxSize" type="number" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="500">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                视频缓存上限 (MB)
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="视频缓存目录的最大容量，超过后自动删除最旧的缓存文件">?</span>
-                            </label>
-                            <input id="cfgVideoCacheMaxSize" type="number" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="1000">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                项目缓存清理
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="当前缓存目录的大小和清理操作">?</span>
-                            </label>
-                            <div class="space-y-2">
-                                <div class="flex gap-2 items-center">
-                                    <span class="text-sm text-muted-foreground w-12">图片:</span>
-                                    <input id="imageCacheSize" readonly class="flex h-9 flex-1 rounded-md border border-input bg-muted px-3 py-2 text-sm" placeholder="0 MB">
-                                    <button onclick="clearImageCache()" class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-destructive text-destructive-foreground hover:bg-destructive/90 h-9 px-3 transition-colors" title="仅清除图片缓存">
-                                        <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                                            <rect x="3" y="3" width="18" height="18" rx="2" ry="2"/>
-                                            <circle cx="9" cy="9" r="2"/>
-                                            <path d="M21 15l-3.086-3.086a2 2 0 0 0-2.828 0L6 21"/>
-                                        </svg>
-                                    </button>
-                                </div>
-                                <div class="flex gap-2 items-center">
-                                    <span class="text-sm text-muted-foreground w-12">视频:</span>
-                                    <input id="videoCacheSize" readonly class="flex h-9 flex-1 rounded-md border border-input bg-muted px-3 py-2 text-sm" placeholder="0 MB">
-                                    <button onclick="clearVideoCache()" class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-destructive text-destructive-foreground hover:bg-destructive/90 h-9 px-3 transition-colors" title="仅清除视频缓存">
-                                        <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                                            <polygon points="23 7 16 12 23 17 23 7"/>
-                                            <rect x="1" y="5" width="15" height="14" rx="2" ry="2"/>
-                                        </svg>
-                                    </button>
-                                </div>
-                                <div class="flex gap-2 items-center">
-                                    <span class="text-sm text-muted-foreground w-12">总计:</span>
-                                    <input id="totalCacheSize" readonly class="flex h-9 flex-1 rounded-md border border-input bg-muted px-3 py-2 text-sm font-medium" placeholder="0 MB">
-                                    <button onclick="clearCache()" class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-destructive text-destructive-foreground hover:bg-destructive/90 h-9 px-3 transition-colors" title="清除所有缓存（图片+视频）">
-                                        <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                                            <path d="M3 6h18"/>
-                                            <path d="M19 6v14a2 2 0 0 1-2 2H7a2 2 0 0 1-2-2V6m3 0V4a2 2 0 0 1 2-2h4a2 2 0 0 1 2 2v2"/>
-                                            <line x1="10" y1="11" x2="10" y2="17"/>
-                                            <line x1="14" y1="11" x2="14" y2="17"/>
-                                        </svg>
-                                    </button>
-                                </div>
-                            </div>
-                        </div>
-                    </div>
-                    <div class="mt-6 flex justify-end">
-                        <button onclick="saveGlobalSettings()" class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-primary text-primary-foreground hover:bg-primary/90 h-9 px-6 transition-colors">保存配置</button>
-                    </div>
-                </div>
-
-                <!-- Grok 配置 -->
-                <div class="rounded-lg border border-border bg-background p-6">
-                    <h3 class="text-lg font-semibold mb-4">Grok 配置</h3>
-                    <div class="space-y-4">
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                API Key
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="API访问密钥。建议填写以提高安全性，可留空">?</span>
-                            </label>
-                            <input id="cfgApiKey" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                Proxy Url
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="HTTP代理服务器地址，用于访问Grok服务，可留空">?</span>
-                            </label>
-                            <input id="cfgProxyUrl" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="http://127.0.0.1:7890">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                CF Clearance
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="Cloudflare安全令牌，用于绕过人机验证，可留空">?</span>
-                            </label>
-                            <input id="cfgCfClearance" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                X Statsig ID
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="Grok反机器人检测的唯一标识符，必填">?</span>
-                            </label>
-                            <input id="cfgStatsigId" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                Filtered Tags
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="过滤Grok响应中的指定标签，多个标签用逗号分隔">?</span>
-                            </label>
-                            <input id="cfgFilteredTags" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="xaiartifact,xai:tool_usage_card">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                临时会话模式
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="选择会话模式：标准模式保留上下文，临时模式每次对话独立">?</span>
-                            </label>
-                            <select id="cfgTemporary" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring">
-                                <option value="false">关闭</option>
-                                <option value="true">开启</option>
-                            </select>
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                首次响应超时 (秒)
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="从请求开始到收到第一个数据块的最大等待时间。超过此时间将结束会话">?</span>
-                            </label>
-                            <input id="cfgStreamFirstResponseTimeout" type="number" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="30">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                流式间隔超时 (秒)
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="流式响应中，两次数据块之间的最大等待时间。超过此时间将结束会话">?</span>
-                            </label>
-                            <input id="cfgStreamChunkTimeout" type="number" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="120">
-                        </div>
-                        <div>
-                            <label class="text-sm font-medium flex items-center gap-1 mb-2">
-                                生成总过程超时 (秒)
-                                <span class="inline-flex items-center justify-center w-3.5 h-3.5 rounded-full border border-muted-foreground text-muted-foreground cursor-help" style="font-size:10px;line-height:1" title="流式响应的总时长限制。设置为0表示不限制总时长">?</span>
-                            </label>
-                            <input id="cfgStreamTotalTimeout" type="number" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm" placeholder="600">
-                        </div>
-                    </div>
-                    <div class="mt-6 flex justify-end">
-                        <button onclick="saveGrokSettings()" class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-primary text-primary-foreground hover:bg-primary/90 h-9 px-6 transition-colors">保存配置</button>
-                    </div>
-                </div>
-            </div>
-        </div>
-    </main>
-
-    <!-- 添加 Token 模态框 -->
-    <div id="addModal" class="hidden fixed inset-0 z-50 bg-black/80 flex items-center justify-center p-4">
-        <div class="bg-background rounded-lg border border-border w-full max-w-2xl max-h-[90vh] overflow-y-auto shadow-xl">
-            <div class="flex items-center justify-between p-5 border-b border-border">
-                <h3 class="text-lg font-semibold">添加 Token</h3>
-                <button onclick="closeAddModal()" class="text-muted-foreground hover:text-foreground transition-colors">
-                    <svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
-                        <line x1="18" y1="6" x2="6" y2="18"/>
-                        <line x1="6" y1="6" x2="18" y2="18"/>
-                    </svg>
-                </button>
-            </div>
-            <div class="p-5 space-y-4">
-                <div class="space-y-2">
-                    <label class="text-sm font-medium">Token 类型</label>
-                    <select id="addTokenType" class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring">
-                        <option value="sso">SSO</option>
-                        <option value="ssoSuper">SuperSSO</option>
-                    </select>
-                </div>
-                <div class="space-y-2">
-                    <label class="text-sm font-medium">Token 列表 <span class="text-muted-foreground">(每行一个)</span></label>
-                    <textarea
-                        id="addTokenList"
-                        rows="12"
-                        placeholder="请输入 Token，每行一个"
-                        class="flex w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring font-mono resize-none"
-                    ></textarea>
-                </div>
-            </div>
-            <div class="flex items-center justify-end gap-3 p-5 border-t border-border bg-muted/30">
-                <button
-                    onclick="closeAddModal()"
-                    class="inline-flex items-center justify-center rounded-md text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring disabled:pointer-events-none disabled:opacity-50 border border-input bg-background hover:bg-accent h-9 px-5"
-                >
-                    取消
-                </button>
-                <button
-                    onclick="submitAddTokens()"
-                    class="inline-flex items-center justify-center rounded-md text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring disabled:pointer-events-none disabled:opacity-50 bg-primary text-primary-foreground hover:bg-primary/90 h-9 px-5"
-                >
-                    添加
-                </button>
-            </div>
-        </div>
+  <main class="mx-auto max-w-7xl px-6 py-6">
+    <!-- Tab 导航 -->
+    <div class="border-b border-border mb-6">
+      <nav class="flex space-x-8">
+        <button onclick="switchTab('tokens')" id="tabTokens"
+          class="tab-btn border-b-2 border-primary text-sm font-medium py-3 px-1">Token 管理</button>
+        <button onclick="switchTab('settings')" id="tabSettings"
+          class="tab-btn border-b-2 border-transparent text-sm font-medium py-3 px-1">Setting 配置</button>
+      </nav>
     </div>
 
-    <script>
-        let allTokens=[],filteredTokens=[],selectedTokens=new Set();
-        const $=(id)=>document.getElementById(id),
-        checkAuth=()=>{const t=localStorage.getItem('adminToken');return t||(location.href='/login',null),t},
-        apiRequest=async(url,opts={})=>{const t=checkAuth();if(!t)return null;const r=await fetch(url,{...opts,headers:{...opts.headers,Authorization:`Bearer ${t}`,'Content-Type':'application/json'}});return r.status===401?(localStorage.removeItem('adminToken'),location.href='/login',null):r},
-        loadStats=async()=>{try{const r=await apiRequest('/api/stats');if(!r)return;const d=await r.json();if(d.success){const s=d.data;$('statTotal').textContent=s.total||0;['Unused','Limited','Expired','Active'].forEach((k,i)=>$(`stat${k}`).textContent=(s.normal?.[k.toLowerCase()]||0)+(s.super?.[k.toLowerCase()]||0))}}catch(e){console.error('加载统计失败:',e)}},
-        calcRemaining=()=>{let n=0,h=0;allTokens.forEach(t=>{if(t.remaining_queries>0)n+=t.remaining_queries;if(t.heavy_remaining_queries>0)h+=t.heavy_remaining_queries});return{normal:n,heavy:h,total:n+h}},
-        loadTokens=async()=>{try{const r=await apiRequest('/api/tokens');if(!r)return;const d=await r.json();d.success&&(allTokens=d.data,filteredTokens=allTokens,selectedTokens.clear(),renderTokens(),updateRemaining())}catch(e){console.error('加载列表失败:',e)}},
-        updateRemaining=()=>{const r=calcRemaining();['Total','Normal','Heavy'].forEach(k=>$(`stat${k}Remaining`).textContent=r[k.toLowerCase()]===0?'-':r[k.toLowerCase()].toLocaleString())}
+    <!-- Token 管理面板 -->
+    <div id="panelTokens">
+      <!-- 统计卡片 -->
+      <div class="grid gap-4 grid-cols-2 md:grid-cols-4 mb-6">
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Token 总数</p>
+          <h3 class="text-xl font-bold" id="statTotal">-</h3>
+        </div>
 
-        const renderTokens=()=>{const tb=$('tokenTableBody'),es=$('emptyState'),ss={'未使用':'bg-muted text-muted-foreground','限流中':'bg-orange-50 text-orange-700 border-orange-200','失效':'bg-destructive/10 text-destructive border-destructive/20','正常':'bg-green-50 text-green-700 border-green-200'},ts={sso:'bg-blue-50 text-blue-700 border-blue-200',ssoSuper:'bg-purple-50 text-purple-700 border-purple-200'},tl={sso:'SSO',ssoSuper:'SuperSSO'};if(!filteredTokens.length){tb.innerHTML='';es.classList.remove('hidden');$('selectAll').checked=false;return updateBatchActions()}es.classList.add('hidden');tb.innerHTML=filteredTokens.map(t=>`<tr class="transition-colors"><td class="py-2.5 px-3 align-middle w-12"><input type="checkbox" class="token-checkbox h-3.5 w-3.5 rounded border border-input focus:ring-1 focus:ring-ring" data-token="${t.token}" data-type="${t.token_type}" ${selectedTokens.has(t.token)?'checked':''} onchange="toggleToken('${t.token}')"></td><td class="py-2.5 px-3 align-middle w-80"><div class="flex items-center gap-2"><span class="font-mono text-xs">${t.token.substring(0,30)}...</span><button onclick="copyToken('${t.token.replace(/'/g,"\\'")}',event)" class="inline-flex items-center justify-center rounded-md transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-accent h-6 w-6" title="复制完整 Token"><svg class="h-3 w-3 text-muted-foreground" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><rect x="9" y="9" width="13" height="13" rx="2" ry="2"/><path d="M5 15H4a2 2 0 0 1-2-2V4a2 2 0 0 1 2-2h9a2 2 0 0 1 2 2v1"/></svg></button></div></td><td class="py-2.5 px-3 align-middle w-20"><span class="inline-flex items-center rounded-full px-1.5 py-0.5 text-xs font-medium border ${ts[t.token_type]}">${tl[t.token_type]}</span></td><td class="py-2.5 px-3 align-middle w-20"><span class="inline-flex items-center rounded-full px-1.5 py-0.5 text-xs font-medium border ${ss[t.status]}">${t.status}</span></td><td class="py-2.5 px-3 align-middle w-20 text-xs tabular-nums">${t.remaining_queries===-1?'-':t.remaining_queries}</td><td class="py-2.5 px-3 align-middle w-20 text-xs tabular-nums">${t.heavy_remaining_queries===-1?'-':t.heavy_remaining_queries}</td><td class="py-2.5 px-3 align-middle w-32 text-xs text-muted-foreground">${t.created_time?new Date(t.created_time).toLocaleString('zh-CN',{dateStyle:'short',timeStyle:'short'}):'-'}</td><td class="py-2.5 px-3 align-middle text-right w-16"><button onclick="deleteToken('${t.token}','${t.token_type}')" class="inline-flex items-center justify-center rounded-md text-xs font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-destructive/10 hover:text-destructive h-7 w-7"><svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><polyline points="3 6 5 6 21 6"/><path d="M19 6v14a2 2 0 0 1-2 2H7a2 2 0 0 1-2-2V6m3 0V4a2 2 0 0 1 2-2h4a2 2 0 0 1 2 2v2"/></svg></button></td></tr>`).join('');updateBatchActions()},
-        toggleToken=t=>selectedTokens[selectedTokens.has(t)?'delete':'add'](t)||updateBatchActions(),
-        toggleSelectAll=()=>{const sa=$('selectAll');sa.checked?filteredTokens.forEach(t=>selectedTokens.add(t.token)):selectedTokens.clear();renderTokens()},
-        updateBatchActions=()=>{const ba=$('batchActions'),sc=$('selectedCount'),c=selectedTokens.size;ba.classList[c>0?'add':'remove']('flex');ba.classList[c>0?'remove':'add']('hidden');c>0&&(sc.textContent=`已选择 ${c} 项`);$('selectAll').checked=filteredTokens.length>0&&c===filteredTokens.length},
-        filterTokens=()=>{const tf=$('filterType').value,sf=$('filterStatus').value;filteredTokens=allTokens.filter(t=>(tf==='all'||t.token_type===tf)&&(sf==='all'||t.status===sf));selectedTokens.clear();renderTokens()},
-        refreshTokens=async()=>{await loadTokens();await loadStats()},
-        openAddModal=()=>$('addModal').classList.remove('hidden'),
-        closeAddModal=()=>{$('addModal').classList.add('hidden');$('addTokenList').value=''},
-        deleteToken=async(t,tt)=>{if(!confirm('确定要删除这个 Token 吗?'))return;try{const r=await apiRequest('/api/tokens/delete',{method:'POST',body:JSON.stringify({tokens:[t],token_type:tt})});if(!r)return;const d=await r.json();d.success?await refreshTokens():alert('删除失败: '+(d.error||'未知错误'))}catch(e){alert('删除失败: '+e.message)}},
-        batchDelete=async()=>{if(!selectedTokens.size||!confirm(`确定要删除选中的 ${selectedTokens.size} 个 Token 吗?此操作不可恢复!`))return;const tbt={sso:[],ssoSuper:[]};document.querySelectorAll('.token-checkbox:checked').forEach(cb=>tbt[cb.dataset.type].push(cb.dataset.token));try{const ps=[];['sso','ssoSuper'].forEach(k=>tbt[k].length&&ps.push(apiRequest('/api/tokens/delete',{method:'POST',body:JSON.stringify({tokens:tbt[k],token_type:k})})));await Promise.all(ps);await refreshTokens()}catch(e){alert('批量删除失败: '+e.message)}},
-        submitAddTokens=async()=>{const tt=$('addTokenType').value,tks=$('addTokenList').value.split('\n').map(t=>t.trim()).filter(t=>t);if(!tks.length)return alert('请输入至少一个 Token');try{const r=await apiRequest('/api/tokens/add',{method:'POST',body:JSON.stringify({tokens:tks,token_type:tt})});if(!r)return;const d=await r.json();d.success?(closeAddModal(),await refreshTokens()):alert('添加失败: '+(d.error||'未知错误'))}catch(e){alert('添加失败: '+e.message)}},
-        copyToken=async(t,e)=>{e.stopPropagation();try{await navigator.clipboard.writeText(t);showToast('Token 已复制到剪贴板','success')}catch(err){console.error('复制失败:',err);showToast('复制失败,请手动复制','error')}}
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Token 正常</p>
+          <h3 class="text-xl font-bold text-green-600" id="statActive">-</h3>
+        </div>
 
-        const exportSelected=()=>{if(!selectedTokens.size)return showToast('请先选择要导出的 Token','error');const sd=allTokens.filter(t=>selectedTokens.has(t.token)),csv=[['Token','类型','状态','普通调用剩余','高级调用剩余','创建时间'].join(','),...sd.map(t=>[`"${t.token}"`,t.token_type==='sso'?'SSO':'SuperSSO',t.status,t.remaining_queries===-1?'未使用':t.remaining_queries,t.heavy_remaining_queries===-1?'未使用':t.heavy_remaining_queries,`"${t.created_time?new Date(t.created_time).toLocaleString('zh-CN'):'-'}"`].join(','))].join('\n'),l=document.createElement('a');l.href=URL.createObjectURL(new Blob(['\uFEFF'+csv],{type:'text/csv;charset=utf-8;'}));l.download=`grok_tokens_${new Date().toISOString().slice(0,10)}.csv`;l.style.display='none';document.body.appendChild(l);l.click();document.body.removeChild(l);URL.revokeObjectURL(l.href);showToast(`已导出 ${selectedTokens.size} 个 Token`,'success')}
-        showToast=(m,t='info')=>{const d=document.createElement('div'),bc={success:'bg-green-600',error:'bg-destructive',info:'bg-primary'};d.className=`fixed bottom-4 right-4 ${bc[t]||bc.info} text-white px-4 py-2.5 rounded-lg shadow-lg text-sm font-medium z-50 animate-slide-up`;d.textContent=m;document.body.appendChild(d);setTimeout(()=>{d.style.opacity='0';d.style.transition='opacity .3s';setTimeout(()=>d.parentNode&&document.body.removeChild(d),300)},2000)}
-        logout=async()=>{if(!confirm('确定要退出登录吗?'))return;try{await apiRequest('/api/logout',{method:'POST'})}catch(e){console.error('登出失败:',e)}finally{localStorage.removeItem('adminToken');location.href='/login'}},
-        switchTab=t=>{['tokens','settings'].forEach(n=>{$(`panel${n.charAt(0).toUpperCase()+n.slice(1)}`).classList[n===t?'remove':'add']('hidden');$(`tab${n.charAt(0).toUpperCase()+n.slice(1)}`).classList[n===t?'add':'remove']('border-primary','text-primary');$(`tab${n.charAt(0).toUpperCase()+n.slice(1)}`).classList[n===t?'remove':'add']('border-transparent','text-muted-foreground')});t==='settings'&&loadSettings()},
-        loadSettings=async()=>{try{const r=await apiRequest('/api/settings');if(!r)return;const d=await r.json();if(d.success){const g=d.data.global,k=d.data.grok;$('cfgAdminUser').value=g.admin_username||'';$('cfgAdminPass').value='';$('cfgLogLevel').value=g.log_level||'DEBUG';$('cfgImageCacheMaxSize').value=g.image_cache_max_size_mb||500;$('cfgVideoCacheMaxSize').value=g.video_cache_max_size_mb||1000;$('cfgImageMode').value=g.image_mode||'url';$('cfgBaseUrl').value=g.base_url||'';$('cfgApiKey').value=k.api_key||'';$('cfgProxyUrl').value=k.proxy_url||'';$('cfgCfClearance').value=k.cf_clearance||'';$('cfgStatsigId').value=k.x_statsig_id||'';$('cfgFilteredTags').value=k.filtered_tags||'';$('cfgTemporary').value=k.temporary!==false?'true':'false';$('cfgStreamChunkTimeout').value=k.stream_chunk_timeout||120;$('cfgStreamFirstResponseTimeout').value=k.stream_first_response_timeout||30;$('cfgStreamTotalTimeout').value=k.stream_total_timeout||600;await loadCacheSize()}}catch(e){console.error('加载配置失败:',e);showToast('加载配置失败','error')}},
-        loadCacheSize=async()=>{try{const r=await apiRequest('/api/cache/size');if(!r)return;const d=await r.json();if(d.success){$('imageCacheSize').value=d.data.image_size||'0 MB';$('videoCacheSize').value=d.data.video_size||'0 MB';$('totalCacheSize').value=d.data.total_size||'0 MB'}}catch(e){console.error('加载缓存大小失败:',e);$('imageCacheSize').value='0 MB';$('videoCacheSize').value='0 MB';$('totalCacheSize').value='0 MB'}},
-        clearImageCache=async()=>{if(!confirm('确定要清理图片缓存吗？此操作将删除所有图片缓存文件！'))return;try{const r=await apiRequest('/api/cache/clear/images',{method:'POST'});if(!r)return;const d=await r.json();d.success?(showToast(`图片缓存清理完成，已删除 ${d.data.deleted_count||0} 个文件`,'success'),await loadCacheSize()):showToast('清理失败: '+(d.error||'未知错误'),'error')}catch(e){showToast('清理失败: '+e.message,'error')}},
-        clearVideoCache=async()=>{if(!confirm('确定要清理视频缓存吗？此操作将删除所有视频缓存文件！'))return;try{const r=await apiRequest('/api/cache/clear/videos',{method:'POST'});if(!r)return;const d=await r.json();d.success?(showToast(`视频缓存清理完成，已删除 ${d.data.deleted_count||0} 个文件`,'success'),await loadCacheSize()):showToast('清理失败: '+(d.error||'未知错误'),'error')}catch(e){showToast('清理失败: '+e.message,'error')}},
-        clearCache=async()=>{if(!confirm('确定要清理缓存吗？此操作将删除 /data/temp 目录中的所有文件！'))return;try{const r=await apiRequest('/api/cache/clear',{method:'POST'});if(!r)return;const d=await r.json();d.success?(showToast(`缓存清理完成，已删除 ${d.data.deleted_count||0} 个文件`,'success'),await loadCacheSize()):showToast('清理失败: '+(d.error||'未知错误'),'error')}catch(e){showToast('清理失败: '+e.message,'error')}},
-        saveGlobalSettings=async()=>{const gc={admin_username:$('cfgAdminUser').value,log_level:$('cfgLogLevel').value,image_cache_max_size_mb:parseInt($('cfgImageCacheMaxSize').value)||500,video_cache_max_size_mb:parseInt($('cfgVideoCacheMaxSize').value)||1000,image_mode:$('cfgImageMode').value,base_url:$('cfgBaseUrl').value};if($('cfgAdminPass').value)gc.admin_password=$('cfgAdminPass').value;try{const r=await apiRequest('/api/settings');if(!r)return;const d=await r.json();if(!d.success)return showToast('加载配置失败','error');const s=await apiRequest('/api/settings',{method:'POST',body:JSON.stringify({global_config:gc,grok_config:d.data.grok})});if(!s)return;const sd=await s.json();sd.success?(showToast('全局配置保存成功','success'),$('cfgAdminPass').value=''):showToast('保存失败: '+(sd.error||'未知错误'),'error')}catch(e){showToast('保存失败: '+e.message,'error')}},
-        saveGrokSettings=async()=>{const kc={api_key:$('cfgApiKey').value,proxy_url:$('cfgProxyUrl').value,cf_clearance:$('cfgCfClearance').value,x_statsig_id:$('cfgStatsigId').value,filtered_tags:$('cfgFilteredTags').value,temporary:$('cfgTemporary').value==='true',stream_chunk_timeout:parseInt($('cfgStreamChunkTimeout').value)||120,stream_first_response_timeout:parseInt($('cfgStreamFirstResponseTimeout').value)||30,stream_total_timeout:parseInt($('cfgStreamTotalTimeout').value)||600};try{const r=await apiRequest('/api/settings');if(!r)return;const d=await r.json();if(!d.success)return showToast('加载配置失败','error');const s=await apiRequest('/api/settings',{method:'POST',body:JSON.stringify({global_config:d.data.global,grok_config:kc})});if(!s)return;const sd=await s.json();sd.success?showToast('Grok配置保存成功','success'):showToast('保存失败: '+(sd.error||'未知错误'),'error')}catch(e){showToast('保存失败: '+e.message,'error')}};
-        const updateHoverCardPosition=(cardElement)=>{const trigger=cardElement.querySelector('.hover-card-trigger'),content=cardElement.querySelector('.hover-card-content');if(!trigger||!content)return;console.log('更新悬浮卡片位置',cardElement,trigger,content);const rect=trigger.getBoundingClientRect(),windowHeight=window.innerHeight,spaceAbove=rect.top,spaceBelow=windowHeight-rect.bottom;content.classList.remove('top','bottom');const originalVisibility=getComputedStyle(content).visibility,originalOpacity=getComputedStyle(content).opacity;content.style.visibility='hidden';content.style.opacity='1';const contentHeight=content.offsetHeight;content.style.visibility=originalVisibility;content.style.opacity=originalOpacity;const position=spaceAbove>contentHeight+10?'top':spaceBelow>contentHeight+10?'bottom':'top';content.classList.add(position);console.log('悬浮卡片位置:',position,'高度:',contentHeight,'上方空间:',spaceAbove,'下方空间:',spaceBelow)},
-        loadStorageMode=async()=>{try{const r=await apiRequest('/api/storage/mode');if(!r)return;const d=await r.json();if(d.success){const mode=d.data.mode;console.log('存储模式:',mode);$('storageModeText').textContent=mode;if(mode==='MYSQL'){$('storageMode').classList.add('bg-blue-50','text-blue-700','border-blue-200');$('storageModeTooltip').textContent='数据库连接模式 - 数据持久化存储，修改配置时可能稍慢但更安全'}else if(mode==='REDIS'){$('storageMode').classList.add('bg-purple-50','text-purple-700','border-purple-200');$('storageModeTooltip').textContent='Redis缓存模式 - 高速内存存储，数据持久化且读写性能极佳'}else{$('storageMode').classList.add('bg-green-50','text-green-700','border-green-200');$('storageModeTooltip').textContent='文件存储模式 - 本地文件存储，读写速度快'};updateHoverCardPosition($('storageMode').closest('.hover-card'))}}catch(e){console.error('加载存储模式失败:',e);$('storageModeText').textContent='FILE';$('storageMode').classList.add('bg-green-50','text-green-700','border-green-200');$('storageModeTooltip').textContent='文件存储模式 - 本地文件存储，读写速度快';updateHoverCardPosition($('storageMode').closest('.hover-card'))}};
-        window.addEventListener('DOMContentLoaded',()=>{checkAuth();loadStorageMode();refreshTokens();setInterval(()=>{loadStats();updateRemaining()},30000);window.addEventListener('resize',()=>{const hoverCard=$('storageMode').closest('.hover-card');hoverCard&&updateHoverCardPosition(hoverCard)});const hoverCard=$('storageMode').closest('.hover-card'),trigger=hoverCard?.querySelector('.hover-card-trigger'),content=hoverCard?.querySelector('.hover-card-content');if(trigger&&content){trigger.addEventListener('mouseenter',()=>{content.style.opacity='1';content.style.visibility='visible'});trigger.addEventListener('mouseleave',()=>{content.style.opacity='0';content.style.visibility='hidden'})}});
-    </script>
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Token 未使用</p>
+          <h3 class="text-xl font-bold text-gray-500" id="statUnused">-</h3>
+        </div>
+
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Token 限流中</p>
+          <h3 class="text-xl font-bold text-orange-600" id="statLimited">-</h3>
+        </div>
+
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Token 失效</p>
+          <h3 class="text-xl font-bold text-destructive" id="statExpired">-</h3>
+        </div>
+
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Chat 总剩余</p>
+          <h3 class="text-xl font-bold" id="statChatRemaining">-</h3>
+        </div>
+
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Image 总剩余</p>
+          <h3 class="text-xl font-bold text-blue-600" id="statImageRemaining">-</h3>
+        </div>
+
+        <div class="rounded-lg border border-border bg-background p-4">
+          <p class="text-sm font-medium text-muted-foreground mb-2">Video 总剩余</p>
+          <h3 class="text-xl font-bold text-purple-600" id="statVideoRemaining">无法统计</h3>
+        </div>
+      </div>
+
+      <!-- Token 列表 -->
+      <div class="rounded-lg border border-border bg-background">
+        <!-- 工具栏 -->
+        <div class="flex items-center justify-between gap-4 p-4 border-b border-border">
+          <div class="flex items-center gap-3 flex-1">
+            <div class="flex items-center gap-2">
+              <select id="filterType" onchange="filterTokens()"
+                class="h-8 px-1 text-sm rounded-md bg-background focus:outline-none focus:ring-1 focus:ring-ring w-[90px]">
+                <option value="all">全部类型</option>
+                <option value="sso">SSO</option>
+                <option value="ssoSuper">SuperSSO</option>
+              </select>
+              <select id="filterStatus" onchange="filterTokens()"
+                class="h-8 px-1 text-sm rounded-md bg-background focus:outline-none focus:ring-1 focus:ring-ring w-[90px]">
+                <option value="all">全部状态</option>
+                <option value="未使用">未使用</option>
+                <option value="限流中">限流中</option>
+                <option value="失效">失效</option>
+                <option value="正常">正常</option>
+              </select>
+              <select id="filterTag" onchange="filterTokens()"
+                class="h-8 px-1 text-sm rounded-md bg-background focus:outline-none focus:ring-1 focus:ring-ring w-[90px]">
+                <option value="all">全部标签</option>
+              </select>
+            </div>
+          </div>
+
+          <div class="flex items-center gap-2">
+            <button onclick="refreshTokens()" class="btn-icon" title="刷新列表">
+              <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                stroke="currentColor" stroke-width="2">
+                <polyline points="23 4 23 10 17 10" />
+                <polyline points="1 20 1 14 7 14" />
+                <path d="M3.51 9a9 9 0 0 1 14.85-3.36L23 10M1 14l4.64 4.36A9 9 0 0 0 20.49 15" />
+              </svg>
+            </button>
+            <div id="batchActions" class="hidden items-center gap-2">
+              <button onclick="exportSelected()" class="btn-icon" title="导出选中项">
+                <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                  stroke="currentColor" stroke-width="2">
+                  <path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4" />
+                  <polyline points="7 10 12 15 17 10" />
+                  <line x1="12" y1="15" x2="12" y2="3" />
+                </svg>
+              </button>
+              <button onclick="batchDelete()" class="btn-icon hover:bg-destructive/10 hover:text-destructive"
+                title="批量删除">
+                <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                  stroke="currentColor" stroke-width="2">
+                  <polyline points="3 6 5 6 21 6" />
+                  <path d="M19 6v14a2 2 0 0 1-2 2H7a2 2 0 0 1-2-2V6m3 0V4a2 2 0 0 1 2-2h4a2 2 0 0 1 2 2v2" />
+                  <line x1="10" y1="11" x2="10" y2="17" />
+                  <line x1="14" y1="11" x2="14" y2="17" />
+                </svg>
+              </button>
+            </div>
+            <button onclick="openAddModal()"
+              class="inline-flex items-center justify-center rounded-md transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring bg-primary text-primary-foreground hover:bg-primary/90 h-8 px-3"
+              title="添加 Token">
+              <svg class="h-4 w-4 mr-2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                stroke="currentColor" stroke-width="2">
+                <line x1="12" y1="5" x2="12" y2="19" />
+                <line x1="5" y1="12" x2="19" y2="12" />
+              </svg>
+              <span class="text-sm font-medium">新增</span>
+            </button>
+          </div>
+        </div>
+
+        <!-- 表格 -->
+        <div class="relative w-full overflow-auto">
+          <table class="w-full text-sm table-fixed">
+            <thead>
+              <tr class="border-b border-border">
+                <th class="h-10 px-3 text-left align-middle font-medium w-12">
+                  <input type="checkbox" id="selectAll" onchange="toggleSelectAll()"
+                    class="h-3.5 w-3.5 rounded border border-input focus:ring-1 focus:ring-ring">
+                </th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-72">Token</th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">类型</th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">状态</th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">普通</th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-20">高级</th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-32">标签</th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-40">备注</th>
+                <th class="h-10 px-3 text-left align-middle text-sm font-medium text-muted-foreground w-32">创建时间</th>
+                <th
+                  class="h-10 px-3 text-center align-middle text-sm font-medium text-muted-foreground w-28 sticky-right">
+                  操作</th>
+              </tr>
+            </thead>
+            <tbody id="tokenTableBody" class="divide-y divide-border">
+              <!-- 动态填充 -->
+            </tbody>
+          </table>
+        </div>
+
+        <div id="emptyState" class="hidden flex flex-col items-center justify-center py-12">
+          <svg class="h-10 w-10 text-muted-foreground/50 mb-3" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"
+            fill="none" stroke="currentColor" stroke-width="2">
+            <rect x="3" y="3" width="18" height="18" rx="2" />
+            <path d="M9 9h6v6H9z" />
+          </svg>
+          <p class="text-sm text-muted-foreground">暂无数据</p>
+        </div>
+      </div>
+    </div>
+
+    <!-- 全局设置面板 -->
+    <div id="panelSettings" class="hidden">
+      <!-- 全局配置区域 -->
+      <div class="mb-8">
+        <div class="flex items-center justify-between mb-6">
+          <h2 class="text-xl font-bold">全局配置</h2>
+          <button onclick="saveGlobalSettings()"
+            class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-secondary text-secondary-foreground hover:bg-black hover:text-white h-9 px-4 transition-colors">保存配置</button>
+        </div>
+        <div class="grid gap-4 lg:grid-cols-3">
+          <!-- 系统设置 -->
+          <div class="rounded-lg border border-border bg-background p-6">
+            <h3 class="text-sm font-semibold mb-4">系统设置</h3>
+            <div class="space-y-4">
+              <div>
+                <label class="cfg-label">登陆账户<span class="help-icon" title="登录管理后台的用户名">?</span></label>
+                <input id="cfgAdminUser" class="cfg-input" placeholder="admin">
+              </div>
+              <div>
+                <label class="cfg-label">登陆密码<span class="help-icon" title="登录管理后台的密码，留空表示不修改当前密码">?</span></label>
+                <input id="cfgAdminPass" type="password" class="cfg-input" placeholder="留空则不修改">
+              </div>
+              <div>
+                <label class="cfg-label">日志级别<span class="help-icon"
+                    title="日志详细程度。DEBUG：最详细 | INFO：一般信息 | WARNING：警告 | ERROR：仅错误">?</span></label>
+                <select id="cfgLogLevel" class="cfg-input">
+                  <option>DEBUG</option>
+                  <option>INFO</option>
+                  <option>WARNING</option>
+                  <option>ERROR</option>
+                </select>
+              </div>
+            </div>
+          </div>
+
+          <!-- 媒体设置 -->
+          <div class="rounded-lg border border-border bg-background p-6">
+            <h3 class="text-sm font-semibold mb-4">媒体设置</h3>
+            <div class="space-y-4">
+              <div>
+                <label class="cfg-label">
+                  图片模式
+                  <span class="help-icon" title="返回图片的方式。URL：图片链接，支持图片缓存 | Base64：base64编码，不支持缓存">?</span>
+                </label>
+                <select id="cfgImageMode" class="cfg-input">
+                  <option value="url">URL链接</option>
+                  <option value="base64">Base64</option>
+                </select>
+              </div>
+              <div>
+                <label class="cfg-label">
+                  服务网址
+                  <span class="help-icon" title="服务器的公网访问地址，用于构建图片URL链接（仅在图片模式为URL时需要）">?</span>
+                </label>
+                <input id="cfgBaseUrl" class="cfg-input" placeholder="http://localhost:8000">
+              </div>
+              <div class="grid grid-cols-2 gap-3">
+                <div>
+                  <label class="cfg-label">
+                    图片缓存 (MB)
+                    <span class="help-icon" title="图片缓存的最大容量（MB），超过后会自动清理旧缓存">?</span>
+                  </label>
+                  <input id="cfgImageCacheMaxSize" type="number" class="cfg-input" placeholder="500">
+                </div>
+                <div>
+                  <label class="cfg-label">
+                    视频缓存 (MB)
+                    <span class="help-icon" title="视频缓存的最大容量（MB），超过后会自动清理旧缓存">?</span>
+                  </label>
+                  <input id="cfgVideoCacheMaxSize" type="number" class="cfg-input" placeholder="1000">
+                </div>
+              </div>
+            </div>
+          </div>
+
+          <!-- 缓存管理 -->
+          <div class="rounded-lg border border-border bg-background p-6">
+            <h3 class="text-sm font-semibold mb-4">缓存管理</h3>
+            <div class="space-y-4">
+              <div>
+                <label class="text-sm font-medium text-muted-foreground mb-2 block">图片缓存</label>
+                <div class="flex gap-2">
+                  <input id="imageCacheSize" readonly
+                    class="flex h-9 flex-1 rounded-md border border-input bg-muted px-3 py-2 text-sm"
+                    placeholder="0 MB">
+                  <button onclick="clearImageCache()"
+                    class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-secondary text-secondary-foreground hover:bg-red-600 hover:text-white h-9 w-9 p-0 transition-colors">
+                    <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                      stroke="currentColor" stroke-width="2">
+                      <path
+                        d="M3 6h18M19 6v14a2 2 0 01-2 2H7a2 2 0 01-2-2V6m3 0V4a2 2 0 012-2h4a2 2 0 012 2v2M10 11v6M14 11v6" />
+                    </svg>
+                  </button>
+                </div>
+              </div>
+              <div>
+                <label class="text-sm font-medium text-muted-foreground mb-2 block">视频缓存</label>
+                <div class="flex gap-2">
+                  <input id="videoCacheSize" readonly
+                    class="flex h-9 flex-1 rounded-md border border-input bg-muted px-3 py-2 text-sm"
+                    placeholder="0 MB">
+                  <button onclick="clearVideoCache()"
+                    class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-secondary text-secondary-foreground hover:bg-red-600 hover:text-white h-9 w-9 p-0 transition-colors">
+                    <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                      stroke="currentColor" stroke-width="2">
+                      <path
+                        d="M3 6h18M19 6v14a2 2 0 01-2 2H7a2 2 0 01-2-2V6m3 0V4a2 2 0 012-2h4a2 2 0 012 2v2M10 11v6M14 11v6" />
+                    </svg>
+                  </button>
+                </div>
+              </div>
+              <div>
+                <label class="text-sm font-medium text-muted-foreground mb-2 block">所有缓存</label>
+                <div class="flex gap-2">
+                  <input id="totalCacheSize" readonly
+                    class="flex h-9 flex-1 rounded-md border border-input bg-muted px-3 py-2 text-sm font-medium"
+                    placeholder="0 MB">
+                  <button onclick="clearCache()"
+                    class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-secondary text-secondary-foreground hover:bg-red-600 hover:text-white h-9 w-9 p-0 transition-colors">
+                    <svg class="h-4 w-4" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+                      stroke="currentColor" stroke-width="2">
+                      <path
+                        d="M3 6h18M19 6v14a2 2 0 01-2 2H7a2 2 0 01-2-2V6m3 0V4a2 2 0 012-2h4a2 2 0 012 2v2M10 11v6M14 11v6" />
+                    </svg>
+                  </button>
+                </div>
+              </div>
+            </div>
+          </div>
+        </div>
+      </div>
+
+      <!-- Grok 配置区域 -->
+      <div>
+        <div class="flex items-center justify-between mb-6">
+          <h2 class="text-xl font-bold">Grok 配置</h2>
+          <button onclick="saveGrokSettings()"
+            class="inline-flex items-center justify-center rounded-md text-sm font-medium bg-secondary text-secondary-foreground hover:bg-black hover:text-white h-9 px-4 transition-colors">保存配置</button>
+        </div>
+        <div class="grid gap-4 lg:grid-cols-3">
+          <!-- 基础设置 -->
+          <div class="rounded-lg border border-border bg-background p-6">
+            <h3 class="text-sm font-semibold mb-4">基础设置</h3>
+            <div class="space-y-4">
+              <div>
+                <label class="cfg-label">
+                  API Key
+                  <span class="help-icon" title="接口调用的身份验证密钥，用于保护API访问安全">?</span>
+                </label>
+                <input id="cfgApiKey" class="cfg-input" placeholder="">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  X Statsig ID
+                  <span class="help-icon" title="Statsig统计ID，用于功能实验和统计分析">?</span>
+                </label>
+                <div class="flex items-center gap-3">
+                  <input id="cfgStatsigId"
+                    class="flex h-9 flex-1 rounded-md border border-input bg-background px-3 py-2 text-sm"
+                    placeholder="">
+                  <label class="inline-flex items-center gap-2 cursor-pointer" title="开启后每次请求自动生成新的 x-statsig-id">
+                    <span class="text-xs text-muted-foreground whitespace-nowrap">动态</span>
+                    <div class="relative">
+                      <input type="checkbox" id="cfgDynamicStatsig" class="sr-only peer">
+                      <div
+                        class="w-9 h-5 bg-gray-200 peer-focus:outline-none peer-focus:ring-2 peer-focus:ring-ring rounded-full peer peer-checked:after:translate-x-full peer-checked:after:border-white after:content-[''] after:absolute after:top-[2px] after:left-[2px] after:bg-white after:border-gray-300 after:border after:rounded-full after:h-4 after:w-4 after:transition-all peer-checked:bg-primary">
+                      </div>
+                    </div>
+                  </label>
+                </div>
+              </div>
+              <div>
+                <label class="cfg-label">
+                  过滤标签
+                  <span class="help-icon" title="需要过滤的响应标签，多个标签用逗号分隔。如：xaiartifact,xai:tool_usage_card">?</span>
+                </label>
+                <input id="cfgFilteredTags" class="cfg-input" placeholder="xaiartifact,xai:tool_usage_card">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  显示思考
+                  <span class="help-icon" title="开启后会显示模型的思考过程（&lt;think&gt;标签内容）；关闭后仅返回最终结果">?</span>
+                </label>
+                <select id="cfgShowThinking" class="cfg-input">
+                  <option value="true">开启</option>
+                  <option value="false">关闭</option>
+                </select>
+              </div>
+              <div>
+                <label class="cfg-label">
+                  临时会话
+                  <span class="help-icon" title="开启后每次对话都创建新会话，不保留历史；关闭后可以继续之前的对话">?</span>
+                </label>
+                <select id="cfgTemporary" class="cfg-input">
+                  <option value="false">关闭</option>
+                  <option value="true">开启</option>
+                </select>
+              </div>
+            </div>
+          </div>
+
+          <!-- 代理设置 -->
+          <div class="rounded-lg border border-border bg-background p-6">
+            <h3 class="text-sm font-semibold mb-4">代理设置</h3>
+            <div class="space-y-4">
+              <div>
+                <label class="cfg-label">
+                  CF Clearance
+                  <span class="help-icon"
+                    title="Cloudflare验证cookie的值部分，用于绕过Cloudflare人机验证。只需输入cf_clearance=后面的值。">?</span>
+                </label>
+                <input id="cfgCfClearance" class="cfg-input" placeholder="">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  Proxy Url (服务代理)
+                  <span class="help-icon"
+                    title="API请求和上传使用的代理。支持 http、https、socks5。格式：socks5://user:pass@host:port">?</span>
+                </label>
+                <input id="cfgProxyUrl" class="cfg-input" placeholder="socks5://username:password@127.0.0.1:7890">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  Proxy Pool URL (代理池API)
+                  <span class="help-icon"
+                    title="代理池API地址，返回单个代理URL。留空则使用上方的固定代理。&#10;&#10;返回格式：纯文本，单行代理地址&#10;示例返回：socks5h://1.2.3.4:1080&#10;支持协议：http://, https://, socks5://, socks5h://&#10;API示例：http://your-api.com/get">?</span>
+                </label>
+                <input id="cfgProxyPoolUrl" class="cfg-input" placeholder="http://your-proxy-api.com/get">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  Proxy Pool Interval (刷新间隔秒)
+                  <span class="help-icon" title="代理池自动刷新间隔，单位秒。建议300-600秒（5-10分钟）">?</span>
+                </label>
+                <input id="cfgProxyPoolInterval" type="number" class="cfg-input" placeholder="300">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  Cache Proxy Url (缓存代理)
+                  <span class="help-icon" title="图片/视频缓存下载专用代理，不设置则使用服务代理。Grok的图片/视频获取接口对IP风控要求不高，可使用便宜的大流量节点">?</span>
+                </label>
+                <input id="cfgCacheProxyUrl" class="cfg-input" placeholder="socks5://username:password@127.0.0.1:7890">
+              </div>
+            </div>
+          </div>
+
+          <!-- 超时设置 -->
+          <div class="rounded-lg border border-border bg-background p-6">
+            <h3 class="text-sm font-semibold mb-4">超时设置</h3>
+            <div class="space-y-4">
+              <div>
+                <label class="cfg-label">
+                  首次响应超时 (秒)
+                  <span class="help-icon" title="等待API首次返回数据的最大时间（秒）。超时后会报错，建议30-60秒">?</span>
+                </label>
+                <input id="cfgStreamFirstResponseTimeout" type="number" class="cfg-input" placeholder="30">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  流式间隔超时 (秒)
+                  <span class="help-icon" title="两次数据块之间的最大间隔时间（秒）。如果超过此时间没有收到新数据则断开，建议60-180秒">?</span>
+                </label>
+                <input id="cfgStreamChunkTimeout" type="number" class="cfg-input" placeholder="120">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  生成总过程超时 (秒)
+                  <span class="help-icon" title="整个对话生成的最大总时长（秒）。适用于超长对话，建议300-900秒">?</span>
+                </label>
+                <input id="cfgStreamTotalTimeout" type="number" class="cfg-input" placeholder="600">
+              </div>
+              <div>
+                <label class="cfg-label">
+                  可重试状态码
+                  <span class="help-icon"
+                    title="遇到这些HTTP状态码时会自动换token重试。多个状态码用逗号分隔。默认：401,429&#10;常见状态码：401(未授权), 429(速率限制), 500(服务器错误), 502(网关错误), 503(服务不可用)">?</span>
+                </label>
+                <input id="cfgRetryStatusCodes" class="cfg-input" placeholder="401,429">
+              </div>
+            </div>
+          </div>
+        </div>
+      </div>
+
+      <!-- 配置提示 -->
+      <div class="mt-8 rounded border border-blue-300 bg-blue-50 px-4 py-3">
+        <div class="text-xs text-gray-800 leading-relaxed">
+          <div class="text-base font-medium text-gray-900 mb-2.5">部分说明</div>
+          <div class="space-y-1.5">
+            <div><span class="font-medium">X Statsig ID：</span>反机器人验证参数。开启"动态 Statsig ID"后会自动生成，固定值将被忽略；关闭则使用上方设置的固定值
+            </div>
+            <div><span class="font-medium">动态 Statsig：</span>开启后每次请求自动生成新的 x-statsig-id，增强请求多样性，推荐开启</div>
+            <div><span class="font-medium">服务网址：</span>图片/视频链接返回时需要拼接您的服务网址（如
+              https://yourdomain.com），若您不使用视频功能且图片使用Base64模式则可留空</div>
+            <div><span class="font-medium">代理设置：</span>服务代理用于访问Grok
+              API和上传图片；缓存代理专门用于下载图片和视频缓存。若仅设置服务代理，缓存将使用相同的代理；若都设置，则分别使用不同的代理</div>
+            <div><span class="font-medium">请求 403：</span>通常是被 CF 拦截了，可采用以下办法之一：1. 更换服务器IP ｜ 2. 配置代理IP ｜ 3.在服务器中访问
+              grok.com 通过 CF 验证后 F12 获取 cf_clearance</div>
+          </div>
+        </div>
+      </div>
+    </div>
+  </main>
+
+  <!-- 编辑信息模态框 -->
+  <div id="editTagsModal" class="hidden fixed inset-0 z-50 bg-black/80 flex items-center justify-center p-4">
+    <div class="bg-background rounded-lg border border-border w-full max-w-md shadow-xl">
+      <div class="flex items-center justify-between p-5 border-b border-border">
+        <h3 class="text-lg font-semibold">编辑信息</h3>
+        <button onclick="closeEditModal()" class="text-muted-foreground hover:text-foreground transition-colors">
+          <svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+            stroke="currentColor" stroke-width="2">
+            <line x1="18" y1="6" x2="6" y2="18" />
+            <line x1="6" y1="6" x2="18" y2="18" />
+          </svg>
+        </button>
+      </div>
+      <div class="p-5 space-y-4">
+        <div class="space-y-2">
+          <label class="text-sm font-medium text-muted-foreground">Token</label>
+          <input id="editTokenInput" readonly
+            class="flex h-9 w-full rounded-md border border-input bg-muted px-3 py-2 text-xs font-mono" placeholder="">
+        </div>
+        <div class="space-y-2">
+          <label class="text-sm font-medium text-muted-foreground">标签 <span
+              class="text-muted-foreground">(多个标签用逗号分隔)</span></label>
+          <input id="editTagsInput" placeholder="例如: 生产环境, 测试用"
+            class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring" />
+          <div id="suggestedTags" class="flex flex-wrap gap-2 mt-2"></div>
+        </div>
+        <div class="space-y-2">
+          <label class="text-sm font-medium text-muted-foreground">备注</label>
+          <textarea id="editNoteInput" rows="3" placeholder="添加备注信息..."
+            class="flex w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring resize-none"></textarea>
+        </div>
+      </div>
+      <div class="flex items-center justify-end gap-3 p-5 border-t border-border bg-muted/30">
+        <button onclick="closeEditModal()"
+          class="inline-flex items-center justify-center rounded-md text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring disabled:pointer-events-none disabled:opacity-50 border border-input bg-background hover:bg-accent h-9 px-5">
+          取消
+        </button>
+        <button onclick="submitEditInfo()"
+          class="inline-flex items-center justify-center rounded-md text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring disabled:pointer-events-none disabled:opacity-50 bg-primary text-primary-foreground hover:bg-primary/90 h-9 px-5">
+          保存
+        </button>
+      </div>
+    </div>
+  </div>
+
+  <!-- 添加 Token 模态框 -->
+  <div id="addModal" class="hidden fixed inset-0 z-50 bg-black/80 flex items-center justify-center p-4">
+    <div class="bg-background rounded-lg border border-border w-full max-w-2xl max-h-[90vh] overflow-y-auto shadow-xl">
+      <div class="flex items-center justify-between p-5 border-b border-border">
+        <h3 class="text-lg font-semibold">添加 Token</h3>
+        <button onclick="closeAddModal()" class="text-muted-foreground hover:text-foreground transition-colors">
+          <svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none"
+            stroke="currentColor" stroke-width="2">
+            <line x1="18" y1="6" x2="6" y2="18" />
+            <line x1="6" y1="6" x2="18" y2="18" />
+          </svg>
+        </button>
+      </div>
+      <div class="p-5 space-y-4">
+        <div class="space-y-2">
+          <label class="text-sm font-medium text-muted-foreground">Token 类型</label>
+          <select id="addTokenType"
+            class="flex h-9 w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring">
+            <option value="sso">SSO</option>
+            <option value="ssoSuper">SuperSSO</option>
+          </select>
+        </div>
+        <div class="space-y-2">
+          <label class="text-sm font-medium text-muted-foreground">Token 列表 <span
+              class="text-muted-foreground">(每行一个)</span></label>
+          <textarea id="addTokenList" rows="12" placeholder="请输入 Token，每行一个"
+            class="flex w-full rounded-md border border-input bg-background px-3 py-2 text-sm focus:outline-none focus:ring-1 focus:ring-ring font-mono resize-none"></textarea>
+        </div>
+      </div>
+      <div class="flex items-center justify-end gap-3 p-5 border-t border-border bg-muted/30">
+        <button onclick="closeAddModal()"
+          class="inline-flex items-center justify-center rounded-md text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring disabled:pointer-events-none disabled:opacity-50 border border-input bg-background hover:bg-accent h-9 px-5">
+          取消
+        </button>
+        <button onclick="submitAddTokens()"
+          class="inline-flex items-center justify-center rounded-md text-sm font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring disabled:pointer-events-none disabled:opacity-50 bg-primary text-primary-foreground hover:bg-primary/90 h-9 px-5">
+          添加
+        </button>
+      </div>
+    </div>
+  </div>
+
+  <script>
+    let allTokens = [], filteredTokens = [], selectedTokens = new Set(), allTagsList = [];
+    const $ = (id) => document.getElementById(id),
+      checkAuth = () => { const t = localStorage.getItem('adminToken'); return t || (location.href = '/login', null), t },
+      apiRequest = async (url, opts = {}) => { const t = checkAuth(); if (!t) return null; const r = await fetch(url, { ...opts, headers: { ...opts.headers, Authorization: `Bearer ${t}`, 'Content-Type': 'application/json' } }); return r.status === 401 ? (localStorage.removeItem('adminToken'), location.href = '/login', null) : r },
+      loadStats = async () => { try { const r = await apiRequest('/api/stats'); if (!r) return; const d = await r.json(); if (d.success) { const s = d.data; $('statTotal').textContent = s.total || 0;['Unused', 'Limited', 'Expired', 'Active'].forEach((k, i) => $(`stat${k}`).textContent = (s.normal?.[k.toLowerCase()] || 0) + (s.super?.[k.toLowerCase()] || 0)) } } catch (e) { console.error('加载统计失败:', e) } },
+      calcRemaining = () => { let n = 0, h = 0; allTokens.forEach(t => { if (t.remaining_queries > 0) n += t.remaining_queries; if (t.heavy_remaining_queries > 0) h += t.heavy_remaining_queries }); return { normal: n, heavy: h, total: n + h } },
+      loadTokens = async () => { try { const r = await apiRequest('/api/tokens'); if (!r) return; const d = await r.json(); d.success && (allTokens = d.data.map(t => ({ ...t, tags: t.tags || [], note: t.note || '' })), filteredTokens = allTokens, selectedTokens.clear(), renderTokens(), updateRemaining(), await loadAllTags()) } catch (e) { console.error('加载列表失败:', e) } },
+      updateRemaining = () => { const r = calcRemaining(); const chatTotal = r.total; const imageTotal = Math.floor(chatTotal / 2); $('statChatRemaining').textContent = chatTotal === 0 ? '-' : chatTotal.toLocaleString(); $('statImageRemaining').textContent = imageTotal === 0 ? '-' : imageTotal.toLocaleString(); $('statVideoRemaining').textContent = '无法统计' }
+
+    const renderTokens = () => { const tb = $('tokenTableBody'), es = $('emptyState'), ss = { '未使用': 'bg-muted text-muted-foreground', '限流中': 'bg-orange-50 text-orange-700 border-orange-200', '失效': 'bg-destructive/10 text-destructive border-destructive/20', '正常': 'bg-green-50 text-green-700 border-green-200' }, ts = { sso: 'bg-blue-50 text-blue-700 border-blue-200', ssoSuper: 'bg-purple-50 text-purple-700 border-purple-200' }, tl = { sso: 'SSO', ssoSuper: 'SuperSSO' }; if (!filteredTokens.length) { tb.innerHTML = ''; es.classList.remove('hidden'); $('selectAll').checked = false; return updateBatchActions() } es.classList.add('hidden'); tb.innerHTML = filteredTokens.map(t => { const tagsHtml = t.tags && t.tags.length ? t.tags.map(tag => `<span class="inline-flex items-center rounded px-1.5 py-0.5 text-xs bg-gray-100 text-gray-700">${tag}</span>`).join(' ') : '<span class="text-xs text-muted-foreground">-</span>'; const noteHtml = t.note && t.note.length ? `<span class="text-xs text-gray-700" title="${t.note}">${t.note.length > 20 ? t.note.substring(0, 20) + '...' : t.note}</span>` : '<span class="text-xs text-muted-foreground">-</span>'; return `<tr class="transition-colors"><td class="py-2.5 px-3 align-middle w-12"><input type="checkbox" class="token-checkbox h-3.5 w-3.5 rounded border border-input focus:ring-1 focus:ring-ring" data-token="${t.token}" data-type="${t.token_type}" ${selectedTokens.has(t.token) ? 'checked' : ''} onchange="toggleToken('${t.token}')"></td><td class="py-2.5 px-3 align-middle w-80"><div class="flex items-center gap-2"><span class="font-mono text-xs">${t.token.substring(0, 30)}...</span><button onclick="copyToken('${t.token.replace(/'/g, "\\'")}',event)" class="inline-flex items-center justify-center rounded-md transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-accent h-6 w-6" title="复制完整 Token"><svg class="h-3 w-3 text-muted-foreground" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><rect x="9" y="9" width="13" height="13" rx="2" ry="2"/><path d="M5 15H4a2 2 0 0 1-2-2V4a2 2 0 0 1 2-2h9a2 2 0 0 1 2 2v1"/></svg></button></div></td><td class="py-2.5 px-3 align-middle w-20"><span class="inline-flex items-center rounded-full px-1.5 py-0.5 text-xs font-medium border ${ts[t.token_type]}">${tl[t.token_type]}</span></td><td class="py-2.5 px-3 align-middle w-20"><span class="inline-flex items-center rounded-full px-1.5 py-0.5 text-xs font-medium border ${ss[t.status]}">${t.status}</span></td><td class="py-2.5 px-3 align-middle w-20 text-xs tabular-nums">${t.remaining_queries === -1 ? '-' : t.remaining_queries}</td><td class="py-2.5 px-3 align-middle w-20 text-xs tabular-nums">${t.heavy_remaining_queries === -1 ? '-' : t.heavy_remaining_queries}</td><td class="py-2.5 px-3 align-middle w-32"><div class="flex flex-wrap gap-1">${tagsHtml}</div></td><td class="py-2.5 px-3 align-middle w-40">${noteHtml}</td><td class="py-2.5 px-3 align-middle w-32 text-xs text-muted-foreground">${t.created_time ? new Date(t.created_time).toLocaleString('zh-CN', { dateStyle: 'short', timeStyle: 'short' }) : '-'}</td><td class="py-2.5 px-3 align-middle text-right w-28 sticky-right"><div class="flex items-center justify-end gap-1"><button onclick="testToken('${t.token}','${t.token_type}')" class="inline-flex items-center justify-center rounded-md text-xs font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-blue-50 hover:text-blue-700 h-7 w-7" title="测试Token"><svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><path d="M22 11.08V12a10 10 0 1 1-5.93-9.14"/><polyline points="22 4 12 14.01 9 11.01"/></svg></button><button onclick="editToken('${t.token}','${t.token_type}')" class="inline-flex items-center justify-center rounded-md text-xs font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-accent hover:text-accent-foreground h-7 w-7" title="编辑信息"><svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><path d="M12 20h9"/><path d="M16.5 3.5a2.121 2.121 0 0 1 3 3L7 19l-4 1 1-4L16.5 3.5z"/></svg></button><button onclick="deleteToken('${t.token}','${t.token_type}')" class="inline-flex items-center justify-center rounded-md text-xs font-medium transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-ring hover:bg-destructive/10 hover:text-destructive h-7 w-7" title="删除"><svg class="h-3.5 w-3.5" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><polyline points="3 6 5 6 21 6"/><path d="M19 6v14a2 2 0 0 1-2 2H7a2 2 0 0 1-2-2V6m3 0V4a2 2 0 0 1 2-2h4a2 2 0 0 1 2 2v2"/></svg></button></div></td></tr>` }).join(''); updateBatchActions() },
+      toggleToken = t => selectedTokens[selectedTokens.has(t) ? 'delete' : 'add'](t) || updateBatchActions(),
+      toggleSelectAll = () => { const sa = $('selectAll'); sa.checked ? filteredTokens.forEach(t => selectedTokens.add(t.token)) : selectedTokens.clear(); renderTokens() },
+      updateBatchActions = () => { const ba = $('batchActions'), sc = $('selectedCount'), c = selectedTokens.size; ba.classList[c > 0 ? 'add' : 'remove']('flex'); ba.classList[c > 0 ? 'remove' : 'add']('hidden'); c > 0 && (sc.textContent = `已选择 ${c} 项`); $('selectAll').checked = filteredTokens.length > 0 && c === filteredTokens.length },
+      filterTokens = () => { const tf = $('filterType').value, sf = $('filterStatus').value, tagf = $('filterTag').value; filteredTokens = allTokens.filter(t => (tf === 'all' || t.token_type === tf) && (sf === 'all' || t.status === sf) && (tagf === 'all' || t.tags && t.tags.includes(tagf))); selectedTokens.clear(); renderTokens() },
+      loadAllTags = async () => { try { const r = await apiRequest('/api/tokens/tags/all'); if (!r) return; const d = await r.json(); if (d.success) { allTagsList = d.data; const tagFilter = $('filterTag'); const currentValue = tagFilter.value; tagFilter.innerHTML = '<option value="all">全部标签</option>' + allTagsList.map(tag => `<option value="${tag}">${tag}</option>`).join(''); tagFilter.value = currentValue } } catch (e) { console.error('加载标签列表失败:', e) } },
+      refreshTokens = async () => { await loadTokens(); await loadStats() },
+      openAddModal = () => $('addModal').classList.remove('hidden'),
+      closeAddModal = () => { $('addModal').classList.add('hidden'); $('addTokenList').value = '' },
+      deleteToken = async (t, tt) => { if (!confirm('确定要删除这个 Token 吗?')) return; try { const r = await apiRequest('/api/tokens/delete', { method: 'POST', body: JSON.stringify({ tokens: [t], token_type: tt }) }); if (!r) return; const d = await r.json(); d.success ? await refreshTokens() : showToast('删除失败: ' + (d.error || '未知错误'), 'error') } catch (e) { showToast('删除失败: ' + e.message, 'error') } },
+      batchDelete = async () => { if (!selectedTokens.size || !confirm(`确定要删除选中的 ${selectedTokens.size} 个 Token 吗?此操作不可恢复!`)) return; const tbt = { sso: [], ssoSuper: [] }; document.querySelectorAll('.token-checkbox:checked').forEach(cb => tbt[cb.dataset.type].push(cb.dataset.token)); try { const ps = [];['sso', 'ssoSuper'].forEach(k => tbt[k].length && ps.push(apiRequest('/api/tokens/delete', { method: 'POST', body: JSON.stringify({ tokens: tbt[k], token_type: k }) }))); await Promise.all(ps); await refreshTokens() } catch (e) { showToast('批量删除失败: ' + e.message, 'error') } },
+      submitAddTokens = async () => { const tt = $('addTokenType').value, tks = $('addTokenList').value.split('\n').map(t => t.trim()).filter(t => t); if (!tks.length) return showToast('请输入至少一个 Token', 'error'); try { const r = await apiRequest('/api/tokens/add', { method: 'POST', body: JSON.stringify({ tokens: tks, token_type: tt }) }); if (!r) return; const d = await r.json(); d.success ? (closeAddModal(), await refreshTokens()) : showToast('添加失败: ' + (d.error || '未知错误'), 'error') } catch (e) { showToast('添加失败: ' + e.message, 'error') } },
+      copyToken = async (t, e) => { e.stopPropagation(); try { await navigator.clipboard.writeText(t); showToast('Token 已复制到剪贴板', 'success') } catch (err) { console.error('复制失败:', err); showToast('复制失败,请手动复制', 'error') } }
+
+    let currentEditToken = '', currentEditTokenType = '';
+    editToken = (token, tokenType) => { currentEditToken = token; currentEditTokenType = tokenType; const tokenData = allTokens.find(t => t.token === token); const currentTags = tokenData?.tags || []; const currentNote = tokenData?.note || ''; $('editTokenInput').value = token.substring(0, 50) + '...'; $('editTagsInput').value = currentTags.join(', '); $('editNoteInput').value = currentNote; const suggestedContainer = $('suggestedTags'); suggestedContainer.innerHTML = ''; if (allTagsList.length > 0) { suggestedContainer.innerHTML = '<div class="text-xs text-muted-foreground mb-1">常用标签:</div>' + allTagsList.map(tag => `<button onclick="addTagToInput('${tag}')" class="inline-flex items-center rounded px-2 py-1 text-xs bg-muted hover:bg-accent transition-colors">${tag}</button>`).join('') } $('editTagsModal').classList.remove('hidden') },
+      closeEditModal = () => { $('editTagsModal').classList.add('hidden'); currentEditToken = ''; currentEditTokenType = '' },
+      addTagToInput = (tag) => { const input = $('editTagsInput'); const currentValue = input.value.trim(); const tags = currentValue ? currentValue.split(',').map(t => t.trim()) : []; if (!tags.includes(tag)) { tags.push(tag); input.value = tags.join(', ') } },
+      submitEditInfo = async () => { if (!currentEditToken) return; const tagsInput = $('editTagsInput').value; const note = $('editNoteInput').value.trim(); const tags = tagsInput.split(',').map(t => t.trim()).filter(t => t); const promises = []; promises.push(apiRequest('/api/tokens/tags', { method: 'POST', body: JSON.stringify({ token: currentEditToken, token_type: currentEditTokenType, tags }) })); promises.push(apiRequest('/api/tokens/note', { method: 'POST', body: JSON.stringify({ token: currentEditToken, token_type: currentEditTokenType, note }) })); try { const results = await Promise.all(promises); const allSuccess = results.every(r => r && r.ok); if (allSuccess) { closeEditModal(); await refreshTokens(); showToast('信息更新成功', 'success') } else { showToast('部分更新失败，请重试', 'error') } } catch (e) { showToast('更新失败: ' + e.message, 'error') } }
+
+    testToken = async (token, tokenType) => { const btn = event.target.closest('button'); const originalHtml = btn.innerHTML; btn.disabled = true; btn.innerHTML = '<svg class="h-3.5 w-3.5 animate-spin" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24"><circle class="opacity-25" cx="12" cy="12" r="10" stroke="currentColor" stroke-width="4"></circle><path class="opacity-75" fill="currentColor" d="M4 12a8 8 0 018-8V0C5.373 0 0 5.373 0 12h4zm2 5.291A7.962 7.962 0 014 12H0c0 3.042 1.135 5.824 3 7.938l3-2.647z"></path></svg>'; try { const r = await apiRequest('/api/tokens/test', { method: 'POST', body: JSON.stringify({ token, token_type: tokenType }) }); if (!r) return; const d = await r.json(); if (d.success && d.data.valid) { showToast(`Token有效！剩余: ${d.data.remaining_queries === -1 ? '无限制' : d.data.remaining_queries}次`, 'success'); await refreshTokens() } else { const errMsgs = { expired: 'Token已失效 (401错误)', blocked: '服务器被block，请稍后再试或更换IP', limited: 'Token已被限流' }; showToast(errMsgs[d.data?.error_type] || 'Token无效或已失效', 'error') } } catch (e) { showToast('测试失败: ' + e.message, 'error') } finally { btn.disabled = false; btn.innerHTML = originalHtml } }
+
+    const exportSelected = () => { if (!selectedTokens.size) return showToast('请先选择要导出的 Token', 'error'); const sd = allTokens.filter(t => selectedTokens.has(t.token)), csv = [['Token', '类型', '状态', '普通调用剩余', '高级调用剩余', '创建时间'].join(','), ...sd.map(t => [`"${t.token}"`, t.token_type === 'sso' ? 'SSO' : 'SuperSSO', t.status, t.remaining_queries === -1 ? '未使用' : t.remaining_queries, t.heavy_remaining_queries === -1 ? '未使用' : t.heavy_remaining_queries, `"${t.created_time ? new Date(t.created_time).toLocaleString('zh-CN') : '-'}"`].join(','))].join('\n'), l = document.createElement('a'); l.href = URL.createObjectURL(new Blob(['\uFEFF' + csv], { type: 'text/csv;charset=utf-8;' })); l.download = `grok_tokens_${new Date().toISOString().slice(0, 10)}.csv`; l.style.display = 'none'; document.body.appendChild(l); l.click(); document.body.removeChild(l); URL.revokeObjectURL(l.href); showToast(`已导出 ${selectedTokens.size} 个 Token`, 'success') }
+    showToast = (m, t = 'info') => { const d = document.createElement('div'), bc = { success: 'bg-green-600', error: 'bg-destructive', info: 'bg-primary' }; d.className = `fixed bottom-4 right-4 ${bc[t] || bc.info} text-white px-4 py-2.5 rounded-lg shadow-lg text-sm font-medium z-50 animate-slide-up`; d.textContent = m; document.body.appendChild(d); setTimeout(() => { d.style.opacity = '0'; d.style.transition = 'opacity .3s'; setTimeout(() => d.parentNode && document.body.removeChild(d), 300) }, 2000) }
+    logout = async () => { if (!confirm('确定要退出登录吗?')) return; try { await apiRequest('/api/logout', { method: 'POST' }) } catch (e) { console.error('登出失败:', e) } finally { localStorage.removeItem('adminToken'); location.href = '/login' } },
+      switchTab = t => { const cap = n => n.charAt(0).toUpperCase() + n.slice(1);['tokens', 'settings'].forEach(n => { const active = n === t; $(`panel${cap(n)}`).classList.toggle('hidden', !active); $(`tab${cap(n)}`).classList.toggle('border-primary', active); $(`tab${cap(n)}`).classList.toggle('text-primary', active); $(`tab${cap(n)}`).classList.toggle('border-transparent', !active); $(`tab${cap(n)}`).classList.toggle('text-muted-foreground', !active) }); t === 'settings' && loadSettings() },
+      updateCacheProxyReadonly = () => { const proxyUrl = $('cfgProxyUrl').value.trim(), cacheProxyInput = $('cfgCacheProxyUrl'); if (proxyUrl) { cacheProxyInput.readOnly = false; cacheProxyInput.classList.remove('bg-muted'); cacheProxyInput.placeholder = 'socks5://username:password@127.0.0.1:7890' } else { cacheProxyInput.readOnly = true; cacheProxyInput.classList.add('bg-muted'); cacheProxyInput.value = ''; cacheProxyInput.placeholder = '设置服务代理后自动启用' } },
+      updateStatsigIdState = () => { const dynamicToggle = $('cfgDynamicStatsig'), statsigInput = $('cfgStatsigId'); if (dynamicToggle.checked) { statsigInput.disabled = true; statsigInput.classList.add('bg-muted', 'text-muted-foreground'); statsigInput.placeholder = '已启用动态生成' } else { statsigInput.disabled = false; statsigInput.classList.remove('bg-muted', 'text-muted-foreground'); statsigInput.placeholder = '' } },
+      loadSettings = async () => { try { const r = await apiRequest('/api/settings'); if (!r) return; const d = await r.json(); if (d.success) { const g = d.data.global, k = d.data.grok; const cfClearance = k.cf_clearance || ''; const cleanCfDisplay = cfClearance.startsWith('cf_clearance=') ? cfClearance.split('cf_clearance=')[1] : cfClearance; $('cfgAdminUser').value = g.admin_username || ''; $('cfgAdminPass').value = ''; $('cfgLogLevel').value = g.log_level || 'DEBUG'; $('cfgImageCacheMaxSize').value = g.image_cache_max_size_mb || 500; $('cfgVideoCacheMaxSize').value = g.video_cache_max_size_mb || 1000; $('cfgImageMode').value = g.image_mode || 'url'; $('cfgBaseUrl').value = g.base_url || ''; $('cfgApiKey').value = k.api_key || ''; $('cfgProxyUrl').value = k.proxy_url || ''; $('cfgProxyPoolUrl').value = k.proxy_pool_url || ''; $('cfgProxyPoolInterval').value = k.proxy_pool_interval || 300; $('cfgCacheProxyUrl').value = k.cache_proxy_url || ''; $('cfgCfClearance').value = cleanCfDisplay; $('cfgStatsigId').value = k.x_statsig_id || ''; $('cfgDynamicStatsig').checked = k.dynamic_statsig !== false; updateStatsigIdState(); $('cfgFilteredTags').value = k.filtered_tags || ''; $('cfgShowThinking').value = k.show_thinking !== false ? 'true' : 'false'; $('cfgTemporary').value = k.temporary !== false ? 'true' : 'false'; $('cfgStreamChunkTimeout').value = k.stream_chunk_timeout || 120; $('cfgStreamFirstResponseTimeout').value = k.stream_first_response_timeout || 30; $('cfgStreamTotalTimeout').value = k.stream_total_timeout || 600; $('cfgRetryStatusCodes').value = (k.retry_status_codes || [401, 429]).join(','); updateCacheProxyReadonly(); await loadCacheSize() } } catch (e) { console.error('加载配置失败:', e); showToast('加载配置失败', 'error') } },
+      loadCacheSize = async () => { try { const r = await apiRequest('/api/cache/size'); if (!r) return; const d = await r.json(); if (d.success) { ['image', 'video', 'total'].forEach(t => $(`${t}CacheSize`).value = d.data[`${t}_size`] || '0 MB') } } catch (e) { console.error('加载缓存大小失败:', e);['image', 'video', 'total'].forEach(t => $(`${t}CacheSize`).value = '0 MB') } },
+      clearCacheByType = async (type, url, msg) => { if (!confirm(msg)) return; try { const r = await apiRequest(url, { method: 'POST' }); if (!r) return; const d = await r.json(); d.success ? (showToast(`${type}缓存清理完成，已删除 ${d.data.deleted_count || 0} 个文件`, 'success'), await loadCacheSize()) : showToast('清理失败: ' + (d.error || '未知错误'), 'error') } catch (e) { showToast('清理失败: ' + e.message, 'error') } },
+      clearImageCache = () => clearCacheByType('图片', '/api/cache/clear/images', '确定要清理图片缓存吗？此操作将删除所有图片缓存文件！'),
+      clearVideoCache = () => clearCacheByType('视频', '/api/cache/clear/videos', '确定要清理视频缓存吗？此操作将删除所有视频缓存文件！'),
+      clearCache = () => clearCacheByType('', '/api/cache/clear', '确定要清理缓存吗？此操作将删除 /data/temp 目录中的所有文件！'),
+      saveGlobalSettings = async () => { const gc = { admin_username: $('cfgAdminUser').value, log_level: $('cfgLogLevel').value, image_cache_max_size_mb: parseInt($('cfgImageCacheMaxSize').value) || 500, video_cache_max_size_mb: parseInt($('cfgVideoCacheMaxSize').value) || 1000, image_mode: $('cfgImageMode').value, base_url: $('cfgBaseUrl').value }; if ($('cfgAdminPass').value) gc.admin_password = $('cfgAdminPass').value; try { const r = await apiRequest('/api/settings'); if (!r) return; const d = await r.json(); if (!d.success) return showToast('加载配置失败', 'error'); const s = await apiRequest('/api/settings', { method: 'POST', body: JSON.stringify({ global_config: gc, grok_config: d.data.grok }) }); if (!s) return; const sd = await s.json(); sd.success ? (showToast('全局配置保存成功', 'success'), $('cfgAdminPass').value = '') : showToast('保存失败: ' + (sd.error || '未知错误'), 'error') } catch (e) { showToast('保存失败: ' + e.message, 'error') } },
+      saveGrokSettings = async () => { const pu = $('cfgProxyUrl').value.trim(), cf = $('cfgCfClearance').value.trim(); const cleanCf = cf.startsWith('cf_clearance=') ? cf.split('cf_clearance=')[1] : cf; const retryCodesStr = $('cfgRetryStatusCodes').value.trim(); const retryCodes = retryCodesStr ? retryCodesStr.split(',').map(c => parseInt(c.trim())).filter(c => !isNaN(c)) : [401, 429]; const kc = { api_key: $('cfgApiKey').value, proxy_url: pu, proxy_pool_url: $('cfgProxyPoolUrl').value.trim(), proxy_pool_interval: parseInt($('cfgProxyPoolInterval').value) || 300, cache_proxy_url: pu ? $('cfgCacheProxyUrl').value : '', cf_clearance: cleanCf, x_statsig_id: $('cfgStatsigId').value, dynamic_statsig: $('cfgDynamicStatsig').checked, filtered_tags: $('cfgFilteredTags').value, show_thinking: $('cfgShowThinking').value === 'true', temporary: $('cfgTemporary').value === 'true', stream_chunk_timeout: parseInt($('cfgStreamChunkTimeout').value) || 120, stream_first_response_timeout: parseInt($('cfgStreamFirstResponseTimeout').value) || 30, stream_total_timeout: parseInt($('cfgStreamTotalTimeout').value) || 600, retry_status_codes: retryCodes }; try { const r = await apiRequest('/api/settings'); if (!r) return; const d = await r.json(); if (!d.success) return showToast('加载配置失败', 'error'); const s = await apiRequest('/api/settings', { method: 'POST', body: JSON.stringify({ global_config: d.data.global, grok_config: kc }) }); if (!s) return; const sd = await s.json(); sd.success ? showToast('Grok配置保存成功', 'success') : showToast('保存失败: ' + (sd.error || '未知错误'), 'error') } catch (e) { showToast('保存失败: ' + e.message, 'error') } };
+    updateHoverCardPosition = c => { const t = c.querySelector('.hover-card-trigger'), ct = c.querySelector('.hover-card-content'); if (!t || !ct) return; const { top, bottom } = t.getBoundingClientRect(), h = window.innerHeight; ct.classList.remove('top', 'bottom'); const { visibility: v, opacity: o } = getComputedStyle(ct); Object.assign(ct.style, { visibility: 'hidden', opacity: '1' }); const ch = ct.offsetHeight; Object.assign(ct.style, { visibility: v, opacity: o }); ct.classList.add(top > ch + 10 ? 'top' : h - bottom > ch + 10 ? 'bottom' : 'top') },
+      loadStorageMode = async () => { const modeConfig = { MYSQL: { classes: ['bg-blue-50', 'text-blue-700', 'border-blue-200'], tooltip: '数据库连接模式 - 数据持久化存储，修改配置时可能稍慢但更安全' }, REDIS: { classes: ['bg-purple-50', 'text-purple-700', 'border-purple-200'], tooltip: 'Redis缓存模式 - 高速内存存储，数据持久化且读写性能极佳' }, FILE: { classes: ['bg-green-50', 'text-green-700', 'border-green-200'], tooltip: '文件存储模式 - 本地文件存储，读写速度快' } }; const applyMode = (mode) => { $('storageModeText').textContent = mode; const config = modeConfig[mode] || modeConfig.FILE; $('storageMode').classList.add(...config.classes); $('storageModeTooltip').textContent = config.tooltip; updateHoverCardPosition($('storageMode').closest('.hover-card')) }; try { const r = await apiRequest('/api/storage/mode'); if (!r) return; const d = await r.json(); d.success && applyMode(d.data.mode) } catch (e) { console.error('加载存储模式失败:', e); applyMode('FILE') } };
+    window.addEventListener('DOMContentLoaded', () => { checkAuth(); loadStorageMode(); refreshTokens(); setInterval(() => { loadStats(); updateRemaining() }, 30000); window.addEventListener('resize', () => { const hoverCard = $('storageMode').closest('.hover-card'); hoverCard && updateHoverCardPosition(hoverCard) }); const hoverCard = $('storageMode').closest('.hover-card'), trigger = hoverCard?.querySelector('.hover-card-trigger'), content = hoverCard?.querySelector('.hover-card-content'); if (trigger && content) { trigger.addEventListener('mouseenter', () => { content.style.opacity = '1'; content.style.visibility = 'visible' }); trigger.addEventListener('mouseleave', () => { content.style.opacity = '0'; content.style.visibility = 'hidden' }) }; $('cfgProxyUrl').addEventListener('input', updateCacheProxyReadonly); $('cfgDynamicStatsig').addEventListener('change', updateStatsigIdState) });
+  </script>
 </body>
+
 </html>
\ No newline at end of file
diff --git a/data/setting.toml b/data/setting.toml
index 5ec8757..6c9fc17 100644
--- a/data/setting.toml
+++ b/data/setting.toml
@@ -1,17 +1,23 @@
 [grok]
 api_key = ""
 proxy_url = ""
-temporary = true
+cache_proxy_url = ""
 cf_clearance = ""
 x_statsig_id = "ZTpUeXBlRXJyb3I6IENhbm5vdCByZWFkIHByb3BlcnRpZXMgb2YgdW5kZWZpbmVkIChyZWFkaW5nICdjaGlsZE5vZGVzJyk="
 filtered_tags = "xaiartifact,xai:tool_usage_card,grok:render"
 stream_chunk_timeout = 120
 stream_total_timeout = 600
 stream_first_response_timeout = 30
+temporary = true
+show_thinking = true
+dynamic_statsig = true
+proxy_pool_url = ""
+proxy_pool_interval = 300
+retry_status_codes = [ 401, 429,]
 
 [global]
 base_url = ""
-log_level = "DEBUG"
+log_level = "INFO"
 image_mode = "url"
 admin_password = "admin"
 admin_username = "admin"
diff --git a/data/temp/video/users-8522ce45-679b-4e0e-a0f7-bb18f434eb6b-generated-15f7113f-5d16-4ff1-bdaa-a2eabd66671c-generated_video.mp4 b/data/temp/video/users-8522ce45-679b-4e0e-a0f7-bb18f434eb6b-generated-15f7113f-5d16-4ff1-bdaa-a2eabd66671c-generated_video.mp4
new file mode 100644
index 0000000..195951e
Binary files /dev/null and b/data/temp/video/users-8522ce45-679b-4e0e-a0f7-bb18f434eb6b-generated-15f7113f-5d16-4ff1-bdaa-a2eabd66671c-generated_video.mp4 differ
diff --git a/docker-entrypoint.sh b/docker-entrypoint.sh
new file mode 100644
index 0000000..b98e9f5
--- /dev/null
+++ b/docker-entrypoint.sh
@@ -0,0 +1,53 @@
+#!/bin/sh
+set -e
+
+# 初始化配置文件（如果不存在）
+echo "[Grok2API] 检查配置文件..."
+
+# 确保数据目录存在
+mkdir -p /app/data/temp/image /app/data/temp/video /app/logs
+
+# 如果 setting.toml 不存在，创建默认配置
+if [ ! -f /app/data/setting.toml ]; then
+    echo "[Grok2API] 初始化 setting.toml..."
+    cat > /app/data/setting.toml << 'EOF'
+[global]
+base_url = "http://localhost:8000"
+log_level = "INFO"
+image_mode = "url"
+admin_password = "admin"
+admin_username = "admin"
+image_cache_max_size_mb = 512
+video_cache_max_size_mb = 1024
+max_upload_concurrency = 20
+max_request_concurrency = 50
+batch_save_interval = 1.0
+batch_save_threshold = 10
+
+[grok]
+api_key = ""
+proxy_url = ""
+cache_proxy_url = ""
+cf_clearance = ""
+x_statsig_id = ""
+dynamic_statsig = true
+filtered_tags = "xaiartifact,xai:tool_usage_card,grok:render"
+stream_chunk_timeout = 120
+stream_total_timeout = 600
+stream_first_response_timeout = 30
+temporary = true
+show_thinking = true
+EOF
+fi
+
+# 如果 token.json 不存在，创建空token文件
+if [ ! -f /app/data/token.json ]; then
+    echo "[Grok2API] 初始化 token.json..."
+    echo '{"ssoNormal": {}, "ssoSuper": {}}' > /app/data/token.json
+fi
+
+echo "[Grok2API] 配置文件检查完成"
+echo "[Grok2API] 启动应用..."
+
+# 执行传入的命令
+exec "$@"
diff --git a/main.py b/main.py
index 3f091dc..cd76815 100644
--- a/main.py
+++ b/main.py
@@ -13,28 +13,77 @@ from app.api.v1.models import router as models_router
 from app.api.v1.images import router as images_router
 from app.api.admin.manage import router as admin_router
 
+# 导入MCP服务器（认证配置在server.py中完成）
+from app.services.mcp import mcp
 
+# 创建MCP的FastAPI应用实例
+# 使用流式HTTP传输，支持高效的双向流式通信
+mcp_app = mcp.http_app(stateless_http=True, transport="streamable-http")
+
+# 2. 定义应用生命周期
 @asynccontextmanager
 async def lifespan(app: FastAPI):
-    """应用生命周期管理"""
-    # 初始化存储管理器
-    await storage_manager.init()
+    """
+    启动顺序:
+    1. 初始化核心服务 (storage, settings, token_manager)
+    2. 异步加载 token 数据
+    3. 启动批量保存任务
+    4. 启动MCP服务生命周期
     
+    关闭顺序 (LIFO):
+    1. 关闭MCP服务生命周期
+    2. 关闭批量保存任务并刷新数据
+    3. 关闭核心服务
+    """
+    # --- 启动过程 ---
+    # 1. 初始化核心服务
+    await storage_manager.init()
+
     # 设置存储到配置和token管理器
     storage = storage_manager.get_storage()
     setting.set_storage(storage)
     token_manager.set_storage(storage)
     
-    # 重新加载配置和token数据（从存储同步）
+    # 2. 重新加载配置
     await setting.reload()
-    token_manager._load_data()
+    logger.info("[Grok2API] 核心服务初始化完成")
     
+    # 2.5. 初始化代理池
+    from app.core.proxy_pool import proxy_pool
+    proxy_url = setting.grok_config.get("proxy_url", "")
+    proxy_pool_url = setting.grok_config.get("proxy_pool_url", "")
+    proxy_pool_interval = setting.grok_config.get("proxy_pool_interval", 300)
+    proxy_pool.configure(proxy_url, proxy_pool_url, proxy_pool_interval)
+    
+    # 3. 异步加载 token 数据
+    await token_manager._load_data()
+    logger.info("[Grok2API] Token数据加载完成")
+    
+    # 4. 启动批量保存任务
+    await token_manager.start_batch_save()
+
+    # 5. 管理MCP服务的生命周期
+    mcp_lifespan_context = mcp_app.lifespan(app)
+    await mcp_lifespan_context.__aenter__()
+    logger.info("[MCP] MCP服务初始化完成")
+
     logger.info("[Grok2API] 应用启动成功")
-    yield
     
-    # 关闭存储连接
-    await storage_manager.close()
-    logger.info("[Grok2API] 应用关闭成功")
+    try:
+        yield
+    finally:
+        # --- 关闭过程 ---
+        # 1. 退出MCP服务的生命周期
+        await mcp_lifespan_context.__aexit__(None, None, None)
+        logger.info("[MCP] MCP服务已关闭")
+        
+        # 2. 关闭批量保存任务并刷新数据
+        await token_manager.shutdown()
+        logger.info("[Token] Token管理器已关闭")
+        
+        # 3. 关闭核心服务
+        await storage_manager.close()
+        logger.info("[Grok2API] 应用关闭成功")
 
 
 # 初始化日志
@@ -44,7 +93,7 @@ logger.info("[Grok2API] 应用正在启动...")
 app = FastAPI(
     title="Grok2API",
     description="Grok API 转换服务",
-    version="1.0.3",
+    version="1.3.1",
     lifespan=lifespan
 )
 
@@ -57,12 +106,12 @@ app.include_router(models_router, prefix="/v1")
 app.include_router(images_router)
 app.include_router(admin_router)
 
-# 挂载静态文件（注意：这个应该在API路由之后，避免拦截API请求）
+# 挂载静态文件
 app.mount("/static", StaticFiles(directory="app/template"), name="template")
 
 @app.get("/")
 async def root():
-    """根路径 - 重定向到登录页"""
+    """根路径"""
     from fastapi.responses import RedirectResponse
     return RedirectResponse(url="/login")
 
@@ -76,7 +125,28 @@ async def health_check():
         "version": "1.0.3"
     }
 
+# 挂载MCP服务器 
+app.mount("", mcp_app)
+
 
 if __name__ == "__main__":
     import uvicorn
-    uvicorn.run(app, host="0.0.0.0", port=8001)
+    import os
+    
+    # 从环境变量读取 worker 数量，默认为 1
+    workers = int(os.getenv("WORKERS", "1"))
+    
+    # 提示多进程模式
+    if workers > 1:
+        logger.info(
+            f"[Grok2API] 多进程模式已启用 (workers={workers})。"
+            f"建议使用 Redis/MySQL 存储以获得最佳性能。"
+        )
+    
+    uvicorn.run(
+        "main:app",  # 使用字符串以支持多worker
+        host="0.0.0.0",
+        port=8001,
+        workers=workers,
+        loop="uvloop" if workers == 1 else "auto"  # 单进程用uvloop，多进程用auto
+    )
\ No newline at end of file
diff --git a/readme.md b/readme.md
index a9d7b16..2e1e76e 100644
--- a/readme.md
+++ b/readme.md
@@ -78,7 +78,7 @@ services:
       # - DATABASE_URL=mysql://user:password@host:3306/grok2api
 
       ## MySQL格式: mysql://user:password@host:port/database
-      ## Redis格式: redis://host:port/db 或 redis://user:password@host:port/db
+      ## Redis格式: redis://host:port/db 或 redis://user:password@host:port/db (SSL: rediss://)
 
 volumes:
   grok_data:
@@ -129,6 +129,11 @@ volumes:
 | POST  | /api/cache/clear/images | 清理图片缓存       | ✅   |
 | POST  | /api/cache/clear/videos | 清理视频缓存       | ✅   |
 | GET   | /api/stats              | 获取统计信息       | ✅   |
+| POST  | /api/tokens/tags        | 更新 Token 标签     | ✅   |
+| POST  | /api/tokens/note        | 更新 Token 备注     | ✅   |
+| POST  | /api/tokens/test        | 测试 Token 可用性   | ✅   |
+| GET   | /api/tokens/tags/all    | 获取所有标签列表    | ✅   |
+| GET   | /api/storage/mode       | 获取存储模式信息    | ✅   |
 
 </details>
 
@@ -138,12 +143,14 @@ volumes:
 
 | 模型名称               | 计次   | 账户类型      | 图像生成/编辑 | 深度思考 | 联网搜索 | 视频生成 |
 |------------------------|--------|--------------|--------------|----------|----------|----------|
-| `grok-3-fast`          | 1      | Basic/Super  | ✅           | ❌       | ✅       | ❌       |
+| `grok-4.1`             | 1      | Basic/Super  | ✅           | ✅       | ✅       | ❌       |
+| `grok-4.1-thinking`    | 1      | Basic/Super  | ✅           | ✅       | ✅       | ❌       |
+| `grok-imagine-0.9`     | -      | Basic/Super  | ✅           | ❌       | ❌       | ✅       |
 | `grok-4-fast`          | 1      | Basic/Super  | ✅           | ✅       | ✅       | ❌       |
 | `grok-4-fast-expert`   | 4      | Basic/Super  | ✅           | ✅       | ✅       | ❌       |
 | `grok-4-expert`        | 4      | Basic/Super  | ✅           | ✅       | ✅       | ❌       |
 | `grok-4-heavy`         | 1      | Super        | ✅           | ✅       | ✅       | ❌       |
-| `grok-imagine-0.9`     | -      | Basic/Super  | ✅           | ❌       | ❌       | ✅       |
+| `grok-3-fast`          | 1      | Basic/Super  | ✅           | ❌       | ✅       | ❌       |
 
 <br>
 
@@ -168,6 +175,7 @@ volumes:
 | cf_clearance               | grok    | 否   | Cloudflare安全令牌                      | ""     |
 | x_statsig_id               | grok    | 是   | 反机器人唯一标识符                      | "ZTpUeXBlRXJyb3I6IENhbm5vdCByZWFkIHByb3BlcnRpZXMgb2YgdW5kZWZpbmVkIChyZWFkaW5nICdjaGlsZE5vZGVzJyk=" |
 | filtered_tags              | grok    | 否   | 过滤响应标签（逗号分隔）                | "xaiartifact,xai:tool_usage_card,grok:render" |
+| show_thinking              | grok    | 否   | 显示思考过程 true(显示)/false(隐藏)     | true   |
 | temporary                  | grok    | 否   | 会话模式 true(临时)/false               | true   |
 
 <br>
diff --git a/requirements.txt b/requirements.txt
index 7b97964..9e51d00 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,11 +1,18 @@
+
 toml==0.10.2
 fastapi==0.119.0
 uvicorn==0.37.0
+uvloop==0.21.0
 python-dotenv==1.1.1
 curl_cffi==0.13.0
 requests==2.32.5
 starlette==0.48.0
 pydantic==2.12.2
 aiofiles==25.1.0
+portalocker==3.0.0
 aiomysql==0.2.0
-redis==6.4.0
\ No newline at end of file
+redis==6.4.0
+fastmcp==2.12.4
+cryptography==46.0.3
+orjson==3.11.4
+aiohttp==3.13.2
\ No newline at end of file
diff --git a/test_concurrency.py b/test_concurrency.py
new file mode 100755
index 0000000..c0089b1
--- /dev/null
+++ b/test_concurrency.py
@@ -0,0 +1,276 @@
+#!/usr/bin/env python3
+"""
+Grok2API 并发性能测试脚本
+
+测试不同并发级别下的API性能表现
+"""
+
+import asyncio
+import aiohttp
+import time
+import statistics
+import argparse
+from datetime import datetime
+from typing import List, Dict, Any
+import json
+
+
+class ConcurrencyTester:
+    """并发测试器"""
+    
+    def __init__(self, base_url: str, api_key: str = None):
+        self.base_url = base_url.rstrip('/')
+        self.api_key = api_key
+        self.results: List[Dict[str, Any]] = []
+    
+    async def test_request(self, session: aiohttp.ClientSession, request_id: int) -> Dict[str, Any]:
+        """发送单个测试请求"""
+        url = f"{self.base_url}/v1/chat/completions"
+        
+        headers = {
+            "Content-Type": "application/json"
+        }
+        if self.api_key:
+            headers["Authorization"] = f"Bearer {self.api_key}"
+        
+        payload = {
+            "model": "grok-3-fast",
+            "messages": [
+                {"role": "user", "content": f"测试请求 #{request_id}，请简短回复OK"}
+            ],
+            "stream": False,
+            "max_tokens": 10
+        }
+        
+        start_time = time.time()
+        
+        try:
+            async with session.post(url, json=payload, headers=headers, timeout=30) as response:
+                status = response.status
+                
+                if status == 200:
+                    data = await response.json()
+                    elapsed = time.time() - start_time
+                    
+                    return {
+                        "id": request_id,
+                        "status": "success",
+                        "http_status": status,
+                        "elapsed": elapsed,
+                        "response_length": len(json.dumps(data))
+                    }
+                else:
+                    elapsed = time.time() - start_time
+                    error_text = await response.text()
+                    
+                    return {
+                        "id": request_id,
+                        "status": "error",
+                        "http_status": status,
+                        "elapsed": elapsed,
+                        "error": error_text[:200]
+                    }
+        
+        except asyncio.TimeoutError:
+            elapsed = time.time() - start_time
+            return {
+                "id": request_id,
+                "status": "timeout",
+                "elapsed": elapsed,
+                "error": "Request timeout"
+            }
+        
+        except Exception as e:
+            elapsed = time.time() - start_time
+            return {
+                "id": request_id,
+                "status": "exception",
+                "elapsed": elapsed,
+                "error": str(e)
+            }
+    
+    async def run_concurrent_test(self, concurrency: int, total_requests: int):
+        """运行并发测试"""
+        print(f"\n{'='*60}")
+        print(f"📊 测试配置：并发数 {concurrency}, 总请求数 {total_requests}")
+        print(f"{'='*60}")
+        
+        connector = aiohttp.TCPConnector(limit=concurrency, limit_per_host=concurrency)
+        timeout = aiohttp.ClientTimeout(total=60)
+        
+        async with aiohttp.ClientSession(connector=connector, timeout=timeout) as session:
+            # 预热
+            print("🔥 预热中...")
+            await self.test_request(session, 0)
+            
+            # 开始测试
+            print(f"🚀 开始并发测试...")
+            start_time = time.time()
+            
+            # 创建任务
+            tasks = []
+            for i in range(1, total_requests + 1):
+                task = asyncio.create_task(self.test_request(session, i))
+                tasks.append(task)
+                
+                # 控制并发数
+                if len(tasks) >= concurrency:
+                    results = await asyncio.gather(*tasks)
+                    self.results.extend(results)
+                    tasks = []
+                    
+                    # 显示进度
+                    print(f"  进度: {i}/{total_requests} ({i/total_requests*100:.1f}%)", end='\r')
+            
+            # 处理剩余任务
+            if tasks:
+                results = await asyncio.gather(*tasks)
+                self.results.extend(results)
+            
+            total_time = time.time() - start_time
+            
+            # 统计和输出
+            self.print_statistics(concurrency, total_requests, total_time)
+    
+    def print_statistics(self, concurrency: int, total_requests: int, total_time: float):
+        """打印统计信息"""
+        success_results = [r for r in self.results if r["status"] == "success"]
+        error_results = [r for r in self.results if r["status"] != "success"]
+        
+        success_count = len(success_results)
+        error_count = len(error_results)
+        
+        if success_results:
+            latencies = [r["elapsed"] for r in success_results]
+            avg_latency = statistics.mean(latencies)
+            min_latency = min(latencies)
+            max_latency = max(latencies)
+            p50_latency = statistics.median(latencies)
+            p95_latency = sorted(latencies)[int(len(latencies) * 0.95)] if len(latencies) > 1 else latencies[0]
+            p99_latency = sorted(latencies)[int(len(latencies) * 0.99)] if len(latencies) > 1 else latencies[0]
+        else:
+            avg_latency = min_latency = max_latency = p50_latency = p95_latency = p99_latency = 0
+        
+        throughput = total_requests / total_time if total_time > 0 else 0
+        
+        print(f"\n\n{'='*60}")
+        print(f"📈 测试结果统计")
+        print(f"{'='*60}")
+        print(f"  测试时间: {total_time:.2f}s")
+        print(f"  总请求数: {total_requests}")
+        print(f"  并发数: {concurrency}")
+        print(f"")
+        print(f"  成功请求: {success_count} ({success_count/total_requests*100:.1f}%)")
+        print(f"  失败请求: {error_count} ({error_count/total_requests*100:.1f}%)")
+        print(f"")
+        print(f"  吞吐量: {throughput:.2f} req/s")
+        print(f"")
+        print(f"  延迟统计:")
+        print(f"    最小: {min_latency*1000:.0f}ms")
+        print(f"    平均: {avg_latency*1000:.0f}ms")
+        print(f"    最大: {max_latency*1000:.0f}ms")
+        print(f"    P50:  {p50_latency*1000:.0f}ms")
+        print(f"    P95:  {p95_latency*1000:.0f}ms")
+        print(f"    P99:  {p99_latency*1000:.0f}ms")
+        
+        # 错误详情
+        if error_results:
+            print(f"\n  ⚠️  错误详情:")
+            error_types = {}
+            for r in error_results:
+                error_type = r.get("status", "unknown")
+                error_types[error_type] = error_types.get(error_type, 0) + 1
+            
+            for error_type, count in error_types.items():
+                print(f"    {error_type}: {count}")
+        
+        print(f"{'='*60}\n")
+        
+        # 性能评级
+        self.print_performance_rating(throughput, avg_latency)
+    
+    def print_performance_rating(self, throughput: float, avg_latency: float):
+        """打印性能评级"""
+        print(f"🎯 性能评级:")
+        
+        # 吞吐量评级
+        if throughput >= 100:
+            rating = "⭐⭐⭐⭐⭐ 优秀"
+        elif throughput >= 60:
+            rating = "⭐⭐⭐⭐ 良好"
+        elif throughput >= 30:
+            rating = "⭐⭐⭐ 中等"
+        elif throughput >= 10:
+            rating = "⭐⭐ 较低"
+        else:
+            rating = "⭐ 需优化"
+        
+        print(f"  吞吐量 ({throughput:.1f} req/s): {rating}")
+        
+        # 延迟评级
+        if avg_latency < 0.5:
+            rating = "⭐⭐⭐⭐⭐ 优秀"
+        elif avg_latency < 1.0:
+            rating = "⭐⭐⭐⭐ 良好"
+        elif avg_latency < 2.0:
+            rating = "⭐⭐⭐ 中等"
+        elif avg_latency < 5.0:
+            rating = "⭐⭐ 较高"
+        else:
+            rating = "⭐ 需优化"
+        
+        print(f"  平均延迟 ({avg_latency*1000:.0f}ms): {rating}")
+        print()
+
+
+async def main():
+    """主函数"""
+    parser = argparse.ArgumentParser(description='Grok2API 并发性能测试')
+    parser.add_argument('--url', default='http://localhost:8001', help='API 基础URL')
+    parser.add_argument('--key', default='', help='API Key（可选）')
+    parser.add_argument('-c', '--concurrency', type=int, default=10, help='并发数')
+    parser.add_argument('-n', '--requests', type=int, default=50, help='总请求数')
+    parser.add_argument('--multi-test', action='store_true', help='运行多级并发测试')
+    
+    args = parser.parse_args()
+    
+    print(f"""
+╔══════════════════════════════════════════════════════════╗
+║          Grok2API 并发性能测试工具                        ║
+╚══════════════════════════════════════════════════════════╝
+
+🔗 测试目标: {args.url}
+🔑 API Key: {'已设置' if args.key else '未设置'}
+⏰ 开始时间: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
+""")
+    
+    tester = ConcurrencyTester(args.url, args.key)
+    
+    if args.multi_test:
+        # 多级并发测试
+        test_configs = [
+            (5, 20),    # 5并发，20请求
+            (10, 50),   # 10并发，50请求
+            (20, 100),  # 20并发，100请求
+            (50, 200),  # 50并发，200请求
+        ]
+        
+        for concurrency, requests in test_configs:
+            tester.results = []  # 清空结果
+            await tester.run_concurrent_test(concurrency, requests)
+            await asyncio.sleep(2)  # 间隔2秒
+    else:
+        # 单次测试
+        await tester.run_concurrent_test(args.concurrency, args.requests)
+    
+    print(f"\n✅ 测试完成！")
+    print(f"⏰ 结束时间: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n")
+
+
+if __name__ == "__main__":
+    try:
+        asyncio.run(main())
+    except KeyboardInterrupt:
+        print("\n\n⚠️  测试被用户中断")
+    except Exception as e:
+        print(f"\n\n❌ 测试失败: {e}")
diff --git a/test_concurrency.sh b/test_concurrency.sh
new file mode 100755
index 0000000..b100e2c
--- /dev/null
+++ b/test_concurrency.sh
@@ -0,0 +1,177 @@
+#!/bin/bash
+
+# Grok2API 并发测试脚本（Shell版本）
+# 使用 curl 和 GNU parallel 进行并发测试
+
+set -e
+
+# 配置
+BASE_URL="${BASE_URL:-http://localhost:8001}"
+API_KEY="${API_KEY:-}"
+CONCURRENCY="${CONCURRENCY:-10}"
+TOTAL_REQUESTS="${TOTAL_REQUESTS:-50}"
+
+# 颜色
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m' # No Color
+
+echo -e "${BLUE}╔══════════════════════════════════════════════════════════╗${NC}"
+echo -e "${BLUE}║          Grok2API 并发性能测试工具 (Shell版)             ║${NC}"
+echo -e "${BLUE}╚══════════════════════════════════════════════════════════╝${NC}"
+echo ""
+echo -e "${GREEN}🔗 测试目标:${NC} $BASE_URL"
+echo -e "${GREEN}🔑 API Key:${NC} ${API_KEY:-(未设置)}"
+echo -e "${GREEN}📊 并发数:${NC} $CONCURRENCY"
+echo -e "${GREEN}📈 总请求数:${NC} $TOTAL_REQUESTS"
+echo ""
+
+# 检查依赖
+if ! command -v curl &> /dev/null; then
+    echo -e "${RED}❌ 错误: 需要安装 curl${NC}"
+    exit 1
+fi
+
+# 创建临时目录
+TMP_DIR=$(mktemp -d)
+trap "rm -rf $TMP_DIR" EXIT
+
+# 单个请求函数
+test_request() {
+    local request_id=$1
+    local start_time=$(date +%s.%N)
+    
+    # 构建请求
+    local headers="Content-Type: application/json"
+    if [ -n "$API_KEY" ]; then
+        headers="${headers}\nAuthorization: Bearer ${API_KEY}"
+    fi
+    
+    local response=$(curl -s -w "\n%{http_code}\n%{time_total}" \
+        -X POST "${BASE_URL}/v1/chat/completions" \
+        -H "Content-Type: application/json" \
+        ${API_KEY:+-H "Authorization: Bearer $API_KEY"} \
+        -d "{
+            \"model\": \"grok-3-fast\",
+            \"messages\": [{\"role\": \"user\", \"content\": \"测试请求 #${request_id}，请简短回复OK\"}],
+            \"stream\": false,
+            \"max_tokens\": 10
+        }" 2>&1)
+    
+    local http_code=$(echo "$response" | tail -n 2 | head -n 1)
+    local time_total=$(echo "$response" | tail -n 1)
+    
+    # 记录结果
+    echo "${request_id},${http_code},${time_total}" >> "$TMP_DIR/results.csv"
+    
+    # 显示进度
+    echo -ne "\r  进度: ${request_id}/${TOTAL_REQUESTS}"
+}
+
+# 导出函数供 parallel 使用
+export -f test_request
+export BASE_URL API_KEY TMP_DIR
+
+# 清空结果文件
+echo "id,status,time" > "$TMP_DIR/results.csv"
+
+echo -e "${YELLOW}🚀 开始并发测试...${NC}"
+START_TIME=$(date +%s.%N)
+
+# 使用 GNU parallel（如果可用），否则使用简单循环
+if command -v parallel &> /dev/null; then
+    seq 1 $TOTAL_REQUESTS | parallel -j $CONCURRENCY test_request {}
+else
+    # 简单的后台任务并发
+    for i in $(seq 1 $TOTAL_REQUESTS); do
+        test_request $i &
+        
+        # 控制并发数
+        if (( i % CONCURRENCY == 0 )); then
+            wait
+        fi
+    done
+    wait
+fi
+
+END_TIME=$(date +%s.%N)
+TOTAL_TIME=$(echo "$END_TIME - $START_TIME" | bc)
+
+echo -e "\n"
+
+# 统计结果
+echo -e "${BLUE}═══════════════════════════════════════════════════════════${NC}"
+echo -e "${BLUE}📈 测试结果统计${NC}"
+echo -e "${BLUE}═══════════════════════════════════════════════════════════${NC}"
+
+# 统计成功/失败
+SUCCESS_COUNT=$(awk -F',' '$2 == 200 {count++} END {print count+0}' "$TMP_DIR/results.csv")
+ERROR_COUNT=$((TOTAL_REQUESTS - SUCCESS_COUNT))
+
+echo -e "  测试时间: ${TOTAL_TIME}s"
+echo -e "  总请求数: ${TOTAL_REQUESTS}"
+echo -e "  并发数: ${CONCURRENCY}"
+echo ""
+echo -e "  成功请求: ${GREEN}${SUCCESS_COUNT}${NC} ($(echo "scale=1; $SUCCESS_COUNT * 100 / $TOTAL_REQUESTS" | bc)%)"
+echo -e "  失败请求: ${RED}${ERROR_COUNT}${NC} ($(echo "scale=1; $ERROR_COUNT * 100 / $TOTAL_REQUESTS" | bc)%)"
+echo ""
+
+# 计算吞吐量
+THROUGHPUT=$(echo "scale=2; $TOTAL_REQUESTS / $TOTAL_TIME" | bc)
+echo -e "  吞吐量: ${GREEN}${THROUGHPUT}${NC} req/s"
+echo ""
+
+# 延迟统计（只统计成功的请求）
+if [ $SUCCESS_COUNT -gt 0 ]; then
+    echo -e "  延迟统计:"
+    
+    # 提取成功请求的延迟时间
+    awk -F',' '$2 == 200 {print $3}' "$TMP_DIR/results.csv" | sort -n > "$TMP_DIR/latencies.txt"
+    
+    MIN=$(head -n 1 "$TMP_DIR/latencies.txt" | awk '{printf "%.0f", $1*1000}')
+    MAX=$(tail -n 1 "$TMP_DIR/latencies.txt" | awk '{printf "%.0f", $1*1000}')
+    AVG=$(awk '{sum+=$1; count++} END {printf "%.0f", sum/count*1000}' "$TMP_DIR/latencies.txt")
+    
+    # P50
+    P50_LINE=$((SUCCESS_COUNT / 2))
+    P50=$(sed -n "${P50_LINE}p" "$TMP_DIR/latencies.txt" | awk '{printf "%.0f", $1*1000}')
+    
+    # P95
+    P95_LINE=$(echo "scale=0; $SUCCESS_COUNT * 0.95 / 1" | bc)
+    P95=$(sed -n "${P95_LINE}p" "$TMP_DIR/latencies.txt" | awk '{printf "%.0f", $1*1000}')
+    
+    # P99
+    P99_LINE=$(echo "scale=0; $SUCCESS_COUNT * 0.99 / 1" | bc)
+    P99=$(sed -n "${P99_LINE}p" "$TMP_DIR/latencies.txt" | awk '{printf "%.0f", $1*1000}')
+    
+    echo -e "    最小: ${MIN}ms"
+    echo -e "    平均: ${AVG}ms"
+    echo -e "    最大: ${MAX}ms"
+    echo -e "    P50:  ${P50}ms"
+    echo -e "    P95:  ${P95}ms"
+    echo -e "    P99:  ${P99}ms"
+fi
+
+echo -e "${BLUE}═══════════════════════════════════════════════════════════${NC}"
+
+# 性能评级
+echo -e "${YELLOW}🎯 性能评级:${NC}"
+
+if (( $(echo "$THROUGHPUT >= 100" | bc -l) )); then
+    RATING="⭐⭐⭐⭐⭐ 优秀"
+elif (( $(echo "$THROUGHPUT >= 60" | bc -l) )); then
+    RATING="⭐⭐⭐⭐ 良好"
+elif (( $(echo "$THROUGHPUT >= 30" | bc -l) )); then
+    RATING="⭐⭐⭐ 中等"
+elif (( $(echo "$THROUGHPUT >= 10" | bc -l) )); then
+    RATING="⭐⭐ 较低"
+else
+    RATING="⭐ 需优化"
+fi
+
+echo -e "  吞吐量 (${THROUGHPUT} req/s): ${RATING}"
+
+echo ""
+echo -e "${GREEN}✅ 测试完成！${NC}"