feat: add fadeInWithBlur animation to Tailwind CSS and update Markdown component

- Introduced a new `fadeInWithBlur` keyframe animation in Tailwind CSS for enhanced visual effects. - Removed inline fade animation styles from the `Markdown` component to streamline rendering. - Updated the `SmoothFade` function to utilize the new animation, improving the user experience during content transitions.
chore: remove @radix-ui/react-slot dependency and update utility functions
2025-09-26 19:34:49 +08:00 · 2025-09-26 19:17:19 +08:00 · 2025-09-26 17:46:51 +08:00
134 changed files with 2493 additions and 5324 deletions
--- a/.github/workflows/auto-i18n.yml
+++ b/.github/workflows/auto-i18n.yml
@@ -2,8 +2,8 @@ name: Auto I18N

 env:
  API_KEY: ${{ secrets.TRANSLATE_API_KEY }}
-  MODEL: ${{ vars.AUTO_I18N_MODEL || 'deepseek/deepseek-v3.1'}}
-  BASE_URL: ${{ vars.AUTO_I18N_BASE_URL || 'https://api.ppinfra.com/openai'}}
+  MODEL: ${{ vars.MODEL || 'deepseek/deepseek-v3.1'}}
+  BASE_URL: ${{ vars.BASE_URL || 'https://api.ppinfra.com/openai'}}

 on:
  pull_request:
@@ -26,7 +26,7 @@ jobs:
          ref: ${{ github.event.pull_request.head.ref }}

      - name: 📦 Setting Node.js
-        uses: actions/setup-node@v5
+        uses: actions/setup-node@v4
        with:
          node-version: 20

--- a/.github/workflows/claude-code-review.yml
+++ b/.github/workflows/claude-code-review.yml
@@ -27,7 +27,7 @@ jobs:

    steps:
      - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
        with:
          fetch-depth: 1

--- a/.github/workflows/claude-translator.yml
+++ b/.github/workflows/claude-translator.yml
@@ -29,7 +29,7 @@ jobs:

    steps:
      - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
        with:
          fetch-depth: 1

--- a/.github/workflows/claude.yml
+++ b/.github/workflows/claude.yml
@@ -37,7 +37,7 @@ jobs:
      actions: read # Required for Claude to read CI results on PRs
    steps:
      - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
        with:
          fetch-depth: 1

--- a/.github/workflows/delete-branch.yml
+++ b/.github/workflows/delete-branch.yml
@@ -12,7 +12,7 @@ jobs:
    if: github.event.pull_request.merged == true && github.event.pull_request.head.repo.full_name == github.repository
    steps:
      - name: Delete merged branch
-        uses: actions/github-script@v8
+        uses: actions/github-script@v7
        with:
          script: |
            github.rest.git.deleteRef({
--- a/.github/workflows/nightly-build.yml
+++ b/.github/workflows/nightly-build.yml
@@ -56,7 +56,7 @@ jobs:
          ref: main

      - name: Install Node.js
-        uses: actions/setup-node@v5
+        uses: actions/setup-node@v4
        with:
          node-version: 20

@@ -99,9 +99,9 @@ jobs:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          NODE_OPTIONS: --max-old-space-size=8192
          MAIN_VITE_CHERRYAI_CLIENT_SECRET: ${{ secrets.MAIN_VITE_CHERRYAI_CLIENT_SECRET }}
-          MAIN_VITE_MINERU_API_KEY: ${{ secrets.MAIN_VITE_MINERU_API_KEY }}
-          RENDERER_VITE_AIHUBMIX_SECRET: ${{ secrets.RENDERER_VITE_AIHUBMIX_SECRET }}
-          RENDERER_VITE_PPIO_APP_SECRET: ${{ secrets.RENDERER_VITE_PPIO_APP_SECRET }}
+          MAIN_VITE_MINERU_API_KEY: ${{ vars.MAIN_VITE_MINERU_API_KEY }}
+          RENDERER_VITE_AIHUBMIX_SECRET: ${{ vars.RENDERER_VITE_AIHUBMIX_SECRET }}
+          RENDERER_VITE_PPIO_APP_SECRET: ${{ vars.RENDERER_VITE_PPIO_APP_SECRET }}

      - name: Build Mac
        if: matrix.os == 'macos-latest'
@@ -110,15 +110,15 @@ jobs:
        env:
          CSC_LINK: ${{ secrets.CSC_LINK }}
          CSC_KEY_PASSWORD: ${{ secrets.CSC_KEY_PASSWORD }}
-          APPLE_ID: ${{ secrets.APPLE_ID }}
-          APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
-          APPLE_TEAM_ID: ${{ secrets.APPLE_TEAM_ID }}
+          APPLE_ID: ${{ vars.APPLE_ID }}
+          APPLE_APP_SPECIFIC_PASSWORD: ${{ vars.APPLE_APP_SPECIFIC_PASSWORD }}
+          APPLE_TEAM_ID: ${{ vars.APPLE_TEAM_ID }}
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          NODE_OPTIONS: --max-old-space-size=8192
          MAIN_VITE_CHERRYAI_CLIENT_SECRET: ${{ secrets.MAIN_VITE_CHERRYAI_CLIENT_SECRET }}
-          MAIN_VITE_MINERU_API_KEY: ${{ secrets.MAIN_VITE_MINERU_API_KEY }}
-          RENDERER_VITE_AIHUBMIX_SECRET: ${{ secrets.RENDERER_VITE_AIHUBMIX_SECRET }}
-          RENDERER_VITE_PPIO_APP_SECRET: ${{ secrets.RENDERER_VITE_PPIO_APP_SECRET }}
+          MAIN_VITE_MINERU_API_KEY: ${{ vars.MAIN_VITE_MINERU_API_KEY }}
+          RENDERER_VITE_AIHUBMIX_SECRET: ${{ vars.RENDERER_VITE_AIHUBMIX_SECRET }}
+          RENDERER_VITE_PPIO_APP_SECRET: ${{ vars.RENDERER_VITE_PPIO_APP_SECRET }}

      - name: Build Windows
        if: matrix.os == 'windows-latest'
@@ -128,9 +128,9 @@ jobs:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          NODE_OPTIONS: --max-old-space-size=8192
          MAIN_VITE_CHERRYAI_CLIENT_SECRET: ${{ secrets.MAIN_VITE_CHERRYAI_CLIENT_SECRET }}
-          MAIN_VITE_MINERU_API_KEY: ${{ secrets.MAIN_VITE_MINERU_API_KEY }}
-          RENDERER_VITE_AIHUBMIX_SECRET: ${{ secrets.RENDERER_VITE_AIHUBMIX_SECRET }}
-          RENDERER_VITE_PPIO_APP_SECRET: ${{ secrets.RENDERER_VITE_PPIO_APP_SECRET }}
+          MAIN_VITE_MINERU_API_KEY: ${{ vars.MAIN_VITE_MINERU_API_KEY }}
+          RENDERER_VITE_AIHUBMIX_SECRET: ${{ vars.RENDERER_VITE_AIHUBMIX_SECRET }}
+          RENDERER_VITE_PPIO_APP_SECRET: ${{ vars.RENDERER_VITE_PPIO_APP_SECRET }}

      - name: Rename artifacts with nightly format
        shell: bash
--- a/.github/workflows/pr-ci.yml
+++ b/.github/workflows/pr-ci.yml
@@ -24,7 +24,7 @@ jobs:
        uses: actions/checkout@v5

      - name: Install Node.js
-        uses: actions/setup-node@v5
+        uses: actions/setup-node@v4
        with:
          node-version: 20

--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -47,7 +47,7 @@ jobs:
          npm version "$VERSION" --no-git-tag-version --allow-same-version

      - name: Install Node.js
-        uses: actions/setup-node@v5
+        uses: actions/setup-node@v4
        with:
          node-version: 20

@@ -86,9 +86,9 @@ jobs:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          NODE_OPTIONS: --max-old-space-size=8192
          MAIN_VITE_CHERRYAI_CLIENT_SECRET: ${{ secrets.MAIN_VITE_CHERRYAI_CLIENT_SECRET }}
-          MAIN_VITE_MINERU_API_KEY: ${{ secrets.MAIN_VITE_MINERU_API_KEY }}
-          RENDERER_VITE_AIHUBMIX_SECRET: ${{ secrets.RENDERER_VITE_AIHUBMIX_SECRET }}
-          RENDERER_VITE_PPIO_APP_SECRET: ${{ secrets.RENDERER_VITE_PPIO_APP_SECRET }}
+          MAIN_VITE_MINERU_API_KEY: ${{ vars.MAIN_VITE_MINERU_API_KEY }}
+          RENDERER_VITE_AIHUBMIX_SECRET: ${{ vars.RENDERER_VITE_AIHUBMIX_SECRET }}
+          RENDERER_VITE_PPIO_APP_SECRET: ${{ vars.RENDERER_VITE_PPIO_APP_SECRET }}

      - name: Build Mac
        if: matrix.os == 'macos-latest'
@@ -98,15 +98,15 @@ jobs:
        env:
          CSC_LINK: ${{ secrets.CSC_LINK }}
          CSC_KEY_PASSWORD: ${{ secrets.CSC_KEY_PASSWORD }}
-          APPLE_ID: ${{ secrets.APPLE_ID }}
-          APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
-          APPLE_TEAM_ID: ${{ secrets.APPLE_TEAM_ID }}
+          APPLE_ID: ${{ vars.APPLE_ID }}
+          APPLE_APP_SPECIFIC_PASSWORD: ${{ vars.APPLE_APP_SPECIFIC_PASSWORD }}
+          APPLE_TEAM_ID: ${{ vars.APPLE_TEAM_ID }}
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          NODE_OPTIONS: --max-old-space-size=8192
          MAIN_VITE_CHERRYAI_CLIENT_SECRET: ${{ secrets.MAIN_VITE_CHERRYAI_CLIENT_SECRET }}
-          MAIN_VITE_MINERU_API_KEY: ${{ secrets.MAIN_VITE_MINERU_API_KEY }}
-          RENDERER_VITE_AIHUBMIX_SECRET: ${{ secrets.RENDERER_VITE_AIHUBMIX_SECRET }}
-          RENDERER_VITE_PPIO_APP_SECRET: ${{ secrets.RENDERER_VITE_PPIO_APP_SECRET }}
+          MAIN_VITE_MINERU_API_KEY: ${{ vars.MAIN_VITE_MINERU_API_KEY }}
+          RENDERER_VITE_AIHUBMIX_SECRET: ${{ vars.RENDERER_VITE_AIHUBMIX_SECRET }}
+          RENDERER_VITE_PPIO_APP_SECRET: ${{ vars.RENDERER_VITE_PPIO_APP_SECRET }}

      - name: Build Windows
        if: matrix.os == 'windows-latest'
@@ -116,9 +116,9 @@ jobs:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          NODE_OPTIONS: --max-old-space-size=8192
          MAIN_VITE_CHERRYAI_CLIENT_SECRET: ${{ secrets.MAIN_VITE_CHERRYAI_CLIENT_SECRET }}
-          MAIN_VITE_MINERU_API_KEY: ${{ secrets.MAIN_VITE_MINERU_API_KEY }}
-          RENDERER_VITE_AIHUBMIX_SECRET: ${{ secrets.RENDERER_VITE_AIHUBMIX_SECRET }}
-          RENDERER_VITE_PPIO_APP_SECRET: ${{ secrets.RENDERER_VITE_PPIO_APP_SECRET }}
+          MAIN_VITE_MINERU_API_KEY: ${{ vars.MAIN_VITE_MINERU_API_KEY }}
+          RENDERER_VITE_AIHUBMIX_SECRET: ${{ vars.RENDERER_VITE_AIHUBMIX_SECRET }}
+          RENDERER_VITE_PPIO_APP_SECRET: ${{ vars.RENDERER_VITE_PPIO_APP_SECRET }}

      - name: Release
        uses: ncipollo/release-action@v1
--- a/.yarn/patches/@ai-sdk-google-npm-2.0.14-376d8b03cc.patch
+++ b/.yarn/patches/@ai-sdk-google-npm-2.0.14-376d8b03cc.patch
@@ -0,0 +1,36 @@
+diff --git a/dist/index.mjs b/dist/index.mjs
+index 110f37ec18c98b1d55ae2b73cc716194e6f9094d..91d0f336b318833c6cee9599fe91370c0ff75323 100644
+--- a/dist/index.mjs
+++ b/dist/index.mjs
+@@ -447,7 +447,10 @@ function convertToGoogleGenerativeAIMessages(prompt, options) {
+ }
+ 
+ // src/get-model-path.ts
+-function getModelPath(modelId) {
+function getModelPath(modelId, baseURL) {
+  if (baseURL?.includes('cherryin')) {
+    return `models/${modelId}`;
+  }
+   return modelId.includes("/") ? modelId : `models/${modelId}`;
+ }
+ 
+@@ -856,7 +859,8 @@ var GoogleGenerativeAILanguageModel = class {
+       rawValue: rawResponse
+     } = await postJsonToApi2({
+       url: `${this.config.baseURL}/${getModelPath(
+-        this.modelId
+        this.modelId,
+        this.config.baseURL
+       )}:generateContent`,
+       headers: mergedHeaders,
+       body: args,
+@@ -962,7 +966,8 @@ var GoogleGenerativeAILanguageModel = class {
+     );
+     const { responseHeaders, value: response } = await postJsonToApi2({
+       url: `${this.config.baseURL}/${getModelPath(
+-        this.modelId
+        this.modelId,
+        this.config.baseURL
+       )}:streamGenerateContent?alt=sse`,
+       headers,
+       body: args,
--- a/.yarn/patches/@ai-sdk-google-npm-2.0.17-fd88491de4.patch
+++ b/.yarn/patches/@ai-sdk-google-npm-2.0.17-fd88491de4.patch
@@ -1,13 +0,0 @@
-diff --git a/dist/index.mjs b/dist/index.mjs
-index b957cb824faa79cf01ba3a504f221870bd8e306a..4d71d30f655775d61537d9d8b73f6e17d41fa67e 100644
--- a/dist/index.mjs
-+++ b/dist/index.mjs
-@@ -452,7 +452,7 @@ function convertToGoogleGenerativeAIMessages(prompt, options) {
- 
- // src/get-model-path.ts
- function getModelPath(modelId) {
-  return modelId.includes("/") ? modelId : `models/${modelId}`;
-+  return modelId?.includes("models/") ? modelId : `models/${modelId}`;
- }
- 
- // src/google-generative-ai-options.ts
--- a/electron-builder.yml
+++ b/electron-builder.yml
@@ -125,17 +125,59 @@ afterSign: scripts/notarize.js
 artifactBuildCompleted: scripts/artifact-build-completed.js
 releaseInfo:
  releaseNotes: |
-    What's New in v1.6.4
+    <!--LANG:en-->
+    🚀 New Features:
+    - Refactored AI core engine for more efficient and stable content generation
+    - Added support for multiple AI model providers: CherryIN, AiOnly
+    - Added API server functionality for external application integration
+    - Added PaddleOCR document recognition for enhanced document processing
+    - Added Anthropic OAuth authentication support
+    - Added data storage space limit notifications
+    - Added font settings for global and code fonts customization
+    - Added auto-copy feature after translation completion
+    - Added keyboard shortcuts: rename topic, edit last message, etc.
+    - Added text attachment preview for viewing file contents in messages
+    - Added custom window control buttons (minimize, maximize, close)
+    - Support for Qwen long-text (qwen-long) and document analysis (qwen-doc) models with native file uploads
+    - Support for Qwen image recognition models (Qwen-Image)
+    - Added iFlow CLI support
+    - Converted knowledge base and web search to tool-calling approach for better flexibility

-    Features:
-    - Providers: add CherryIN provider
-    - Notes: Add right-click context menu to create notes and folders
-    - Mini App: Add search functionality in mini app page
-    - Update Dialog: Add updating dialog in renderer process
-    - Mini App: Remove some mini apps
+    🎨 UI Improvements & Bug Fixes:
+    - Integrated HeroUI and Tailwind CSS framework
+    - Optimized message notification styles with unified toast component
+    - Moved free models to bottom with fixed position for easier access
+    - Refactored quick panel and input bar tools for smoother operation
+    - Optimized responsive design for navbar and sidebar
+    - Improved scrollbar component with horizontal scrolling support
+    - Fixed multiple translation issues: paste handling, file processing, state management
+    - Various UI optimizations and bug fixes
+    <!--LANG:zh-CN-->
+    🚀 新功能：
+    - 重构 AI 核心引擎，提供更高效稳定的内容生成
+    - 新增多个 AI 模型提供商支持：CherryIN、AiOnly
+    - 新增 API 服务器功能，支持外部应用集成
+    - 新增 PaddleOCR 文档识别，增强文档处理能力
+    - 新增 Anthropic OAuth 认证支持
+    - 新增数据存储空间限制提醒
+    - 新增字体设置，支持全局字体和代码字体自定义
+    - 新增翻译完成后自动复制功能
+    - 新增键盘快捷键：重命名主题、编辑最后一条消息等
+    - 新增文本附件预览，可查看消息中的文件内容
+    - 新增自定义窗口控制按钮（最小化、最大化、关闭）
+    - 支持通义千问长文本（qwen-long）和文档分析（qwen-doc）模型，原生文件上传
+    - 支持通义千问图像识别模型（Qwen-Image）
+    - 新增 iFlow CLI 支持
+    - 知识库和网页搜索转换为工具调用方式，提升灵活性

-    Bug Fixes:
-    - Fix reasoning block insertion order - now inserts before content block
-    - Fix knowledge base deletion and web search RAG errors
-    - Fix Qwen model URL configuration
+    🎨 界面改进与问题修复：
+    - 集成 HeroUI 和 Tailwind CSS 框架
+    - 优化消息通知样式，统一 toast 组件
+    - 免费模型移至底部固定位置，便于访问
+    - 重构快捷面板和输入栏工具，操作更流畅
+    - 优化导航栏和侧边栏响应式设计
+    - 改进滚动条组件，支持水平滚动
+    - 修复多个翻译问题：粘贴处理、文件处理、状态管理
+    - 各种界面优化和问题修复
+    <!--LANG:END-->

--- a/electron.vite.config.ts
+++ b/electron.vite.config.ts
@@ -34,10 +34,6 @@ export default defineConfig({
        output: {
          manualChunks: undefined, // 彻底禁用代码分割 - 返回 null 强制单文件打包
          inlineDynamicImports: true // 内联所有动态导入，这是关键配置
-        },
-        onwarn(warning, warn) {
-          if (warning.code === 'COMMONJS_VARIABLE_IN_ESM') return
-          warn(warning)
        }
      },
      sourcemap: isDev
@@ -115,10 +111,6 @@ export default defineConfig({
          selectionToolbar: resolve(__dirname, 'src/renderer/selectionToolbar.html'),
          selectionAction: resolve(__dirname, 'src/renderer/selectionAction.html'),
          traceWindow: resolve(__dirname, 'src/renderer/traceWindow.html')
-        },
-        onwarn(warning, warn) {
-          if (warning.code === 'COMMONJS_VARIABLE_IN_ESM') return
-          warn(warning)
        }
      }
    },
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
  "name": "CherryStudio",
-  "version": "1.6.4",
+  "version": "1.6.1",
  "private": true,
  "description": "A powerful AI assistant for producer.",
  "main": "./out/main/index.js",
@@ -97,10 +97,10 @@
    "@agentic/exa": "^7.3.3",
    "@agentic/searxng": "^7.3.3",
    "@agentic/tavily": "^7.3.3",
-    "@ai-sdk/amazon-bedrock": "^3.0.29",
-    "@ai-sdk/google-vertex": "^3.0.33",
-    "@ai-sdk/mistral": "^2.0.17",
-    "@ai-sdk/perplexity": "^2.0.11",
+    "@ai-sdk/amazon-bedrock": "^3.0.21",
+    "@ai-sdk/google-vertex": "^3.0.27",
+    "@ai-sdk/mistral": "^2.0.14",
+    "@ai-sdk/perplexity": "^2.0.9",
    "@ant-design/v5-patch-for-react-19": "^1.0.3",
    "@anthropic-ai/sdk": "^0.41.0",
    "@anthropic-ai/vertex-sdk": "patch:@anthropic-ai/vertex-sdk@npm%3A0.11.4#~/.yarn/patches/@anthropic-ai-vertex-sdk-npm-0.11.4-c19cb41edb.patch",
@@ -215,7 +215,7 @@
    "@viz-js/lang-dot": "^1.0.5",
    "@viz-js/viz": "^3.14.0",
    "@xyflow/react": "^12.4.4",
-    "ai": "^5.0.59",
+    "ai": "^5.0.44",
    "antd": "patch:antd@npm%3A5.27.0#~/.yarn/patches/antd-npm-5.27.0-aa91c36546.patch",
    "archiver": "^7.0.1",
    "async-mutex": "^0.5.0",
@@ -238,7 +238,7 @@
    "docx": "^9.0.2",
    "dompurify": "^3.2.6",
    "dotenv-cli": "^7.4.2",
-    "electron": "37.6.0",
+    "electron": "37.4.0",
    "electron-builder": "26.0.15",
    "electron-devtools-installer": "^3.2.0",
    "electron-store": "^8.2.0",
@@ -369,7 +369,7 @@
    "undici": "6.21.2",
    "vite": "npm:rolldown-vite@latest",
    "tesseract.js@npm:*": "patch:tesseract.js@npm%3A6.0.1#~/.yarn/patches/tesseract.js-npm-6.0.1-2562a7e46d.patch",
-    "@ai-sdk/google@npm:2.0.17": "patch:@ai-sdk/google@npm%3A2.0.17#~/.yarn/patches/@ai-sdk-google-npm-2.0.17-fd88491de4.patch"
+    "@ai-sdk/google@npm:2.0.14": "patch:@ai-sdk/google@npm%3A2.0.14#~/.yarn/patches/@ai-sdk-google-npm-2.0.14-376d8b03cc.patch"
  },
  "packageManager": "yarn@4.9.1",
  "lint-staged": {
--- a/packages/aiCore/package.json
+++ b/packages/aiCore/package.json
@@ -1,6 +1,6 @@
 {
  "name": "@cherrystudio/ai-core",
-  "version": "1.0.1",
+  "version": "1.0.0-alpha.18",
  "description": "Cherry Studio AI Core - Unified AI Provider Interface Based on Vercel AI SDK",
  "main": "dist/index.js",
  "module": "dist/index.mjs",
@@ -36,14 +36,15 @@
    "ai": "^5.0.26"
  },
  "dependencies": {
-    "@ai-sdk/anthropic": "^2.0.22",
-    "@ai-sdk/azure": "^2.0.42",
-    "@ai-sdk/deepseek": "^1.0.20",
-    "@ai-sdk/openai": "^2.0.42",
-    "@ai-sdk/openai-compatible": "^1.0.19",
+    "@ai-sdk/anthropic": "^2.0.17",
+    "@ai-sdk/azure": "^2.0.30",
+    "@ai-sdk/deepseek": "^1.0.17",
+    "@ai-sdk/google": "patch:@ai-sdk/google@npm%3A2.0.14#~/.yarn/patches/@ai-sdk-google-npm-2.0.14-376d8b03cc.patch",
+    "@ai-sdk/openai": "^2.0.30",
+    "@ai-sdk/openai-compatible": "^1.0.17",
    "@ai-sdk/provider": "^2.0.0",
-    "@ai-sdk/provider-utils": "^3.0.10",
-    "@ai-sdk/xai": "^2.0.23",
+    "@ai-sdk/provider-utils": "^3.0.9",
+    "@ai-sdk/xai": "^2.0.18",
    "zod": "^4.1.5"
  },
  "devDependencies": {
--- a/packages/aiCore/src/core/plugins/built-in/toolUsePlugin/promptToolUsePlugin.ts
+++ b/packages/aiCore/src/core/plugins/built-in/toolUsePlugin/promptToolUsePlugin.ts
@@ -261,39 +261,22 @@ export const createPromptToolUsePlugin = (config: PromptToolUseConfig = {}) => {
        return params
      }

-      // 分离 provider-defined 和其他类型的工具
-      const providerDefinedTools: ToolSet = {}
-      const promptTools: ToolSet = {}
+      context.mcpTools = params.tools

-      for (const [toolName, tool] of Object.entries(params.tools as ToolSet)) {
-        if (tool.type === 'provider-defined') {
-          // provider-defined 类型的工具保留在 tools 参数中
-          providerDefinedTools[toolName] = tool
-        } else {
-          // 其他工具转换为 prompt 模式
-          promptTools[toolName] = tool
-        }
-      }
-
-      // 只有当有非 provider-defined 工具时才保存到 context
-      if (Object.keys(promptTools).length > 0) {
-        context.mcpTools = promptTools
-      }
-
-      // 构建系统提示符（只包含非 provider-defined 工具）
+      // 构建系统提示符
      const userSystemPrompt = typeof params.system === 'string' ? params.system : ''
-      const systemPrompt = buildSystemPrompt(userSystemPrompt, promptTools)
+      const systemPrompt = buildSystemPrompt(userSystemPrompt, params.tools)
      let systemMessage: string | null = systemPrompt
      if (config.createSystemMessage) {
        // 🎯 如果用户提供了自定义处理函数，使用它
        systemMessage = config.createSystemMessage(systemPrompt, params, context)
      }

-      // 保留 provider-defined tools，移除其他 tools
+      // 移除 tools，改为 prompt 模式
      const transformedParams = {
        ...params,
        ...(systemMessage ? { system: systemMessage } : {}),
-        tools: Object.keys(providerDefinedTools).length > 0 ? providerDefinedTools : undefined
+        tools: undefined
      }
      context.originalParams = transformedParams
      return transformedParams
@@ -302,9 +285,8 @@ export const createPromptToolUsePlugin = (config: PromptToolUseConfig = {}) => {
      let textBuffer = ''
      // let stepId = ''

-      // 如果没有需要 prompt 模式处理的工具，直接返回原始流
      if (!context.mcpTools) {
-        return new TransformStream()
+        throw new Error('No tools available')
      }

      // 从 context 中获取或初始化 usage 累加器
--- a/packages/aiCore/src/core/plugins/built-in/webSearchPlugin/helper.ts
+++ b/packages/aiCore/src/core/plugins/built-in/webSearchPlugin/helper.ts
@@ -1,7 +1,6 @@
 import { anthropic } from '@ai-sdk/anthropic'
 import { google } from '@ai-sdk/google'
 import { openai } from '@ai-sdk/openai'
-import { InferToolInput, InferToolOutput } from 'ai'

 import { ProviderOptionsMap } from '../../../options/types'
 import { OpenRouterSearchConfig } from './openrouter'
@@ -59,31 +58,24 @@ export const DEFAULT_WEB_SEARCH_CONFIG: WebSearchPluginConfig = {

 export type WebSearchToolOutputSchema = {
  // Anthropic 工具 - 手动定义
-  anthropic: InferToolOutput<ReturnType<typeof anthropic.tools.webSearch_20250305>>
+  anthropicWebSearch: Array<{
+    url: string
+    title: string
+    pageAge: string | null
+    encryptedContent: string
+    type: string
+  }>

  // OpenAI 工具 - 基于实际输出
-  // TODO: 上游定义不规范,是unknown
-  // openai: InferToolOutput<ReturnType<typeof openai.tools.webSearch>>
-  openai: {
-    status: 'completed' | 'failed'
-  }
-  'openai-chat': {
+  openaiWebSearch: {
    status: 'completed' | 'failed'
  }
+
  // Google 工具
-  // TODO: 上游定义不规范,是unknown
-  // google: InferToolOutput<ReturnType<typeof google.tools.googleSearch>>
-  google: {
+  googleSearch: {
    webSearchQueries?: string[]
    groundingChunks?: Array<{
      web?: { uri: string; title: string }
    }>
  }
 }
-
-export type WebSearchToolInputSchema = {
-  anthropic: InferToolInput<ReturnType<typeof anthropic.tools.webSearch_20250305>>
-  openai: InferToolInput<ReturnType<typeof openai.tools.webSearch>>
-  google: InferToolInput<ReturnType<typeof google.tools.googleSearch>>
-  'openai-chat': InferToolInput<ReturnType<typeof openai.tools.webSearchPreview>>
-}
--- a/packages/shared/IpcChannel.ts
+++ b/packages/shared/IpcChannel.ts
@@ -5,8 +5,8 @@ export enum IpcChannel {
  App_SetLanguage = 'app:set-language',
  App_SetEnableSpellCheck = 'app:set-enable-spell-check',
  App_SetSpellCheckLanguages = 'app:set-spell-check-languages',
+  App_ShowUpdateDialog = 'app:show-update-dialog',
  App_CheckForUpdate = 'app:check-for-update',
-  App_QuitAndInstall = 'app:quit-and-install',
  App_Reload = 'app:reload',
  App_Quit = 'app:quit',
  App_Info = 'app:info',
@@ -34,7 +34,6 @@ export enum IpcChannel {
  App_GetBinaryPath = 'app:get-binary-path',
  App_InstallUvBinary = 'app:install-uv-binary',
  App_InstallBunBinary = 'app:install-bun-binary',
-  App_InstallOvmsBinary = 'app:install-ovms-binary',
  App_LogToMain = 'app:log-to-main',
  App_SaveData = 'app:save-data',
  App_GetDiskInfo = 'app:get-disk-info',
@@ -221,7 +220,6 @@ export enum IpcChannel {
  // system
  System_GetDeviceType = 'system:getDeviceType',
  System_GetHostname = 'system:getHostname',
-  System_GetCpuName = 'system:getCpuName',

  // DevTools
  System_ToggleDevTools = 'system:toggleDevTools',
@@ -229,6 +227,7 @@ export enum IpcChannel {
  // events
  BackupProgress = 'backup-progress',
  ThemeUpdated = 'theme:updated',
+  UpdateDownloadedCancelled = 'update-downloaded-cancelled',
  RestoreProgress = 'restore-progress',
  UpdateError = 'update-error',
  UpdateAvailable = 'update-available',
@@ -331,15 +330,6 @@ export enum IpcChannel {
  // OCR
  OCR_ocr = 'ocr:ocr',

-  // OVMS
-  Ovms_AddModel = 'ovms:add-model',
-  Ovms_StopAddModel = 'ovms:stop-addmodel',
-  Ovms_GetModels = 'ovms:get-models',
-  Ovms_IsRunning = 'ovms:is-running',
-  Ovms_GetStatus = 'ovms:get-status',
-  Ovms_RunOVMS = 'ovms:run-ovms',
-  Ovms_StopOVMS = 'ovms:stop-ovms',
-
  // CherryAI
  Cherryai_GetSignature = 'cherryai:get-signature'
 }
--- a/packages/shared/config/constant.ts
+++ b/packages/shared/config/constant.ts
@@ -217,8 +217,7 @@ export enum codeTools {
  claudeCode = 'claude-code',
  geminiCli = 'gemini-cli',
  openaiCodex = 'openai-codex',
-  iFlowCli = 'iflow-cli',
-  githubCopilotCli = 'github-copilot-cli'
+  iFlowCli = 'iflow-cli'
 }

 export enum terminalApps {
--- a/resources/scripts/download.js
+++ b/resources/scripts/download.js
@@ -1,7 +1,5 @@
 const https = require('https')
 const fs = require('fs')
-const path = require('path')
-const { execSync } = require('child_process')

 /**
 * Downloads a file from a URL with redirect handling
@@ -34,39 +32,4 @@ async function downloadWithRedirects(url, destinationPath) {
  })
 }

-/**
- * Downloads a file using PowerShell Invoke-WebRequest command
- * @param {string} url The URL to download from
- * @param {string} destinationPath The path to save the file to
- * @returns {Promise<boolean>} Promise that resolves to true if download succeeds
- */
-async function downloadWithPowerShell(url, destinationPath) {
-  return new Promise((resolve, reject) => {
-    try {
-      // Only support windows platform for PowerShell download
-      if (process.platform !== 'win32') {
-        return reject(new Error('PowerShell download is only supported on Windows'))
-      }
-
-      const outputDir = path.dirname(destinationPath)
-      fs.mkdirSync(outputDir, { recursive: true })
-
-      // PowerShell command to download the file with progress disabled for faster download
-      const psCommand = `powershell -Command "$ProgressPreference = 'SilentlyContinue'; Invoke-WebRequest '${url}' -OutFile '${destinationPath}'"`
-
-      console.log(`Downloading with PowerShell: ${url}`)
-      execSync(psCommand, { stdio: 'inherit' })
-
-      if (fs.existsSync(destinationPath)) {
-        console.log(`Download completed: ${destinationPath}`)
-        resolve(true)
-      } else {
-        reject(new Error('Download failed: File not found after download'))
-      }
-    } catch (error) {
-      reject(new Error(`PowerShell download failed: ${error.message}`))
-    }
-  })
-}
-
-module.exports = { downloadWithRedirects, downloadWithPowerShell }
+module.exports = { downloadWithRedirects }
--- a/resources/scripts/install-ovms.js
+++ b/resources/scripts/install-ovms.js
@@ -1,177 +0,0 @@
-const fs = require('fs')
-const path = require('path')
-const os = require('os')
-const { execSync } = require('child_process')
-const { downloadWithPowerShell } = require('./download')
-
-// Base URL for downloading OVMS binaries
-const OVMS_PKG_NAME = 'ovms250911.zip'
-const OVMS_RELEASE_BASE_URL = [`https://gitcode.com/gcw_ggDjjkY3/kjfile/releases/download/download/${OVMS_PKG_NAME}`]
-
-/**
- * Downloads and extracts the OVMS binary for the specified platform
- */
-async function downloadOvmsBinary() {
-  // Create output directory structure - OVMS goes into its own subdirectory
-  const csDir = path.join(os.homedir(), '.cherrystudio')
-
-  // Ensure directories exist
-  fs.mkdirSync(csDir, { recursive: true })
-
-  const csOvmsDir = path.join(csDir, 'ovms')
-  // Delete existing OVMS directory if it exists
-  if (fs.existsSync(csOvmsDir)) {
-    fs.rmSync(csOvmsDir, { recursive: true })
-  }
-
-  const tempdir = os.tmpdir()
-  const tempFilename = path.join(tempdir, 'ovms.zip')
-
-  // Try each URL until one succeeds
-  let downloadSuccess = false
-  let lastError = null
-
-  for (let i = 0; i < OVMS_RELEASE_BASE_URL.length; i++) {
-    const downloadUrl = OVMS_RELEASE_BASE_URL[i]
-    console.log(`Attempting download from URL ${i + 1}/${OVMS_RELEASE_BASE_URL.length}: ${downloadUrl}`)
-
-    try {
-      console.log(`Downloading OVMS from ${downloadUrl} to ${tempFilename}...`)
-
-      // Try PowerShell download first, fallback to Node.js download if it fails
-      await downloadWithPowerShell(downloadUrl, tempFilename)
-
-      // If we get here, download was successful
-      downloadSuccess = true
-      console.log(`Successfully downloaded from: ${downloadUrl}`)
-      break
-    } catch (error) {
-      console.warn(`Download failed from ${downloadUrl}: ${error.message}`)
-      lastError = error
-
-      // Clean up failed download file if it exists
-      if (fs.existsSync(tempFilename)) {
-        try {
-          fs.unlinkSync(tempFilename)
-        } catch (cleanupError) {
-          console.warn(`Failed to clean up temporary file: ${cleanupError.message}`)
-        }
-      }
-
-      // Continue to next URL if this one failed
-      if (i < OVMS_RELEASE_BASE_URL.length - 1) {
-        console.log(`Trying next URL...`)
-      }
-    }
-  }
-
-  // Check if any download succeeded
-  if (!downloadSuccess) {
-    console.error(`All download URLs failed. Last error: ${lastError?.message || 'Unknown error'}`)
-    return 103
-  }
-
-  try {
-    console.log(`Extracting to ${csDir}...`)
-
-    // Use tar.exe to extract the ZIP file
-    console.log(`Extracting OVMS to ${csDir}...`)
-    execSync(`tar -xf ${tempFilename} -C ${csDir}`, { stdio: 'inherit' })
-    console.log(`OVMS extracted to ${csDir}`)
-
-    // Clean up temporary file
-    fs.unlinkSync(tempFilename)
-    console.log(`Installation directory: ${csDir}`)
-  } catch (error) {
-    console.error(`Error installing OVMS: ${error.message}`)
-    if (fs.existsSync(tempFilename)) {
-      fs.unlinkSync(tempFilename)
-    }
-
-    // Check if ovmsDir is empty and remove it if so
-    try {
-      const ovmsDir = path.join(csDir, 'ovms')
-      const files = fs.readdirSync(ovmsDir)
-      if (files.length === 0) {
-        fs.rmSync(ovmsDir, { recursive: true })
-        console.log(`Removed empty directory: ${ovmsDir}`)
-      }
-    } catch (cleanupError) {
-      console.warn(`Warning: Failed to clean up directory: ${cleanupError.message}`)
-      return 105
-    }
-
-    return 104
-  }
-
-  return 0
-}
-
-/**
- * Get the CPU Name and ID
- */
-function getCpuInfo() {
-  const cpuInfo = {
-    name: '',
-    id: ''
-  }
-
-  // Use PowerShell to get CPU information
-  try {
-    const psCommand = `powershell -Command "Get-CimInstance -ClassName Win32_Processor | Select-Object Name, DeviceID | ConvertTo-Json"`
-    const psOutput = execSync(psCommand).toString()
-    const cpuData = JSON.parse(psOutput)
-
-    if (Array.isArray(cpuData)) {
-      cpuInfo.name = cpuData[0].Name || ''
-      cpuInfo.id = cpuData[0].DeviceID || ''
-    } else {
-      cpuInfo.name = cpuData.Name || ''
-      cpuInfo.id = cpuData.DeviceID || ''
-    }
-  } catch (error) {
-    console.error(`Failed to get CPU info: ${error.message}`)
-  }
-
-  return cpuInfo
-}
-
-/**
- * Main function to install OVMS
- */
-async function installOvms() {
-  const platform = os.platform()
-  console.log(`Detected platform: ${platform}`)
-
-  const cpuName = getCpuInfo().name
-  console.log(`CPU Name: ${cpuName}`)
-
-  // Check if CPU name contains "Ultra"
-  if (!cpuName.toLowerCase().includes('intel') || !cpuName.toLowerCase().includes('ultra')) {
-    console.error('OVMS installation requires an Intel(R) Core(TM) Ultra CPU.')
-    return 101
-  }
-
-  // only support windows
-  if (platform !== 'win32') {
-    console.error('OVMS installation is only supported on Windows.')
-    return 102
-  }
-
-  return await downloadOvmsBinary()
-}
-
-// Run the installation
-installOvms()
-  .then((retcode) => {
-    if (retcode === 0) {
-      console.log('OVMS installation successful')
-    } else {
-      console.error('OVMS installation failed')
-    }
-    process.exit(retcode)
-  })
-  .catch((error) => {
-    console.error('OVMS installation failed:', error)
-    process.exit(100)
-  })
--- a/src/main/ipc.ts
+++ b/src/main/ipc.ts
@@ -35,7 +35,6 @@ import NotificationService from './services/NotificationService'
 import * as NutstoreService from './services/NutstoreService'
 import ObsidianVaultService from './services/ObsidianVaultService'
 import { ocrService } from './services/ocr/OcrService'
-import OvmsManager from './services/OvmsManager'
 import { proxyManager } from './services/ProxyManager'
 import { pythonService } from './services/PythonService'
 import { FileServiceManager } from './services/remotefile/FileServiceManager'
@@ -82,7 +81,6 @@ const obsidianVaultService = new ObsidianVaultService()
 const vertexAIService = VertexAIService.getInstance()
 const memoryService = MemoryService.getInstance()
 const dxtService = new DxtService()
-const ovmsManager = new OvmsManager()

 export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) {
  const appUpdater = new AppUpdater()
@@ -132,7 +130,7 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) {
  ipcMain.handle(IpcChannel.Open_Website, (_, url: string) => shell.openExternal(url))

  // Update
-  ipcMain.handle(IpcChannel.App_QuitAndInstall, () => appUpdater.quitAndInstall())
+  ipcMain.handle(IpcChannel.App_ShowUpdateDialog, () => appUpdater.showUpdateDialog(mainWindow))

  // language
  ipcMain.handle(IpcChannel.App_SetLanguage, (_, language) => {
@@ -434,7 +432,6 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) {
  // system
  ipcMain.handle(IpcChannel.System_GetDeviceType, () => (isMac ? 'mac' : isWin ? 'windows' : 'linux'))
  ipcMain.handle(IpcChannel.System_GetHostname, () => require('os').hostname())
-  ipcMain.handle(IpcChannel.System_GetCpuName, () => require('os').cpus()[0].model)
  ipcMain.handle(IpcChannel.System_ToggleDevTools, (e) => {
    const win = BrowserWindow.fromWebContents(e.sender)
    win && win.webContents.toggleDevTools()
@@ -713,7 +710,6 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) {
  ipcMain.handle(IpcChannel.App_GetBinaryPath, (_, name: string) => getBinaryPath(name))
  ipcMain.handle(IpcChannel.App_InstallUvBinary, () => runInstallScript('install-uv.js'))
  ipcMain.handle(IpcChannel.App_InstallBunBinary, () => runInstallScript('install-bun.js'))
-  ipcMain.handle(IpcChannel.App_InstallOvmsBinary, () => runInstallScript('install-ovms.js'))

  //copilot
  ipcMain.handle(IpcChannel.Copilot_GetAuthMessage, CopilotService.getAuthMessage.bind(CopilotService))
@@ -845,17 +841,6 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) {
    ocrService.ocr(file, provider)
  )

-  // OVMS
-  ipcMain.handle(IpcChannel.Ovms_AddModel, (_, modelName: string, modelId: string, modelSource: string, task: string) =>
-    ovmsManager.addModel(modelName, modelId, modelSource, task)
-  )
-  ipcMain.handle(IpcChannel.Ovms_StopAddModel, () => ovmsManager.stopAddModel())
-  ipcMain.handle(IpcChannel.Ovms_GetModels, () => ovmsManager.getModels())
-  ipcMain.handle(IpcChannel.Ovms_IsRunning, () => ovmsManager.initializeOvms())
-  ipcMain.handle(IpcChannel.Ovms_GetStatus, () => ovmsManager.getOvmsStatus())
-  ipcMain.handle(IpcChannel.Ovms_RunOVMS, () => ovmsManager.runOvms())
-  ipcMain.handle(IpcChannel.Ovms_StopOVMS, () => ovmsManager.stopOvms())
-
  // CherryAI
  ipcMain.handle(IpcChannel.Cherryai_GetSignature, (_, params) => generateSignature(params))
 }
--- a/src/main/services/AppUpdater.ts
+++ b/src/main/services/AppUpdater.ts
@@ -1,15 +1,17 @@
 import { loggerService } from '@logger'
 import { isWin } from '@main/constant'
 import { getIpCountry } from '@main/utils/ipService'
+import { locales } from '@main/utils/locales'
 import { generateUserAgent } from '@main/utils/systemInfo'
 import { FeedUrl, UpgradeChannel } from '@shared/config/constant'
 import { IpcChannel } from '@shared/IpcChannel'
 import { CancellationToken, UpdateInfo } from 'builder-util-runtime'
-import { app, net } from 'electron'
+import { app, BrowserWindow, dialog, net } from 'electron'
 import { AppUpdater as _AppUpdater, autoUpdater, Logger, NsisUpdater, UpdateCheckResult } from 'electron-updater'
 import path from 'path'
 import semver from 'semver'

+import icon from '../../../build/icon.png?asset'
 import { configManager } from './ConfigManager'
 import { windowService } from './WindowService'

@@ -24,6 +26,7 @@ const LANG_MARKERS = {

 export default class AppUpdater {
  autoUpdater: _AppUpdater = autoUpdater
+  private releaseInfo: UpdateInfo | undefined
  private cancellationToken: CancellationToken = new CancellationToken()
  private updateCheckResult: UpdateCheckResult | null = null

@@ -63,6 +66,7 @@ export default class AppUpdater {
    autoUpdater.on('update-downloaded', (releaseInfo: UpdateInfo) => {
      const processedReleaseInfo = this.processReleaseInfo(releaseInfo)
      windowService.getMainWindow()?.webContents.send(IpcChannel.UpdateDownloaded, processedReleaseInfo)
+      this.releaseInfo = processedReleaseInfo
      logger.info('update downloaded', processedReleaseInfo)
    })

@@ -243,9 +247,37 @@ export default class AppUpdater {
    }
  }

-  public quitAndInstall() {
-    app.isQuitting = true
-    setImmediate(() => autoUpdater.quitAndInstall())
+  public async showUpdateDialog(mainWindow: BrowserWindow) {
+    if (!this.releaseInfo) {
+      return
+    }
+    const locale = locales[configManager.getLanguage()]
+    const { update: updateLocale } = locale.translation
+
+    let detail = this.formatReleaseNotes(this.releaseInfo.releaseNotes)
+    if (detail === '') {
+      detail = updateLocale.noReleaseNotes
+    }
+
+    dialog
+      .showMessageBox({
+        type: 'info',
+        title: updateLocale.title,
+        icon,
+        message: updateLocale.message.replace('{{version}}', this.releaseInfo.version),
+        detail,
+        buttons: [updateLocale.later, updateLocale.install],
+        defaultId: 1,
+        cancelId: 0
+      })
+      .then(({ response }) => {
+        if (response === 1) {
+          app.isQuitting = true
+          setImmediate(() => autoUpdater.quitAndInstall())
+        } else {
+          mainWindow.webContents.send(IpcChannel.UpdateDownloadedCancelled)
+        }
+      })
  }

  /**
@@ -317,9 +349,38 @@ export default class AppUpdater {

    return processedInfo
  }
+
+  /**
+   * Format release notes for display
+   * @param releaseNotes - Release notes in various formats
+   * @returns Formatted string for display
+   */
+  private formatReleaseNotes(releaseNotes: string | ReleaseNoteInfo[] | null | undefined): string {
+    if (!releaseNotes) {
+      return ''
+    }
+
+    if (typeof releaseNotes === 'string') {
+      // Check if it contains multi-language markers
+      if (this.hasMultiLanguageMarkers(releaseNotes)) {
+        return this.parseMultiLangReleaseNotes(releaseNotes)
+      }
+      return releaseNotes
+    }
+
+    if (Array.isArray(releaseNotes)) {
+      return releaseNotes.map((note) => note.note).join('\n')
+    }
+
+    return ''
+  }
 }
 interface GithubReleaseInfo {
  draft: boolean
  prerelease: boolean
  tag_name: string
 }
+interface ReleaseNoteInfo {
+  readonly version: string
+  readonly note: string | null
+}
--- a/src/main/services/CodeToolsService.ts
+++ b/src/main/services/CodeToolsService.ts
@@ -31,10 +31,7 @@ interface VersionInfo {

 class CodeToolsService {
  private versionCache: Map<string, { version: string; timestamp: number }> = new Map()
-  private terminalsCache: {
-    terminals: TerminalConfig[]
-    timestamp: number
-  } | null = null
+  private terminalsCache: { terminals: TerminalConfig[]; timestamp: number } | null = null
  private customTerminalPaths: Map<string, string> = new Map() // Store user-configured terminal paths
  private readonly CACHE_DURATION = 1000 * 60 * 30 // 30 minutes cache
  private readonly TERMINALS_CACHE_DURATION = 1000 * 60 * 5 // 5 minutes cache for terminals
@@ -85,8 +82,6 @@ class CodeToolsService {
        return '@qwen-code/qwen-code'
      case codeTools.iFlowCli:
        return '@iflow-ai/iflow-cli'
-      case codeTools.githubCopilotCli:
-        return '@github/copilot'
      default:
        throw new Error(`Unsupported CLI tool: ${cliTool}`)
    }
@@ -104,8 +99,6 @@ class CodeToolsService {
        return 'qwen'
      case codeTools.iFlowCli:
        return 'iflow'
-      case codeTools.githubCopilotCli:
-        return 'copilot'
      default:
        throw new Error(`Unsupported CLI tool: ${cliTool}`)
    }
@@ -151,9 +144,7 @@ class CodeToolsService {
        case terminalApps.powershell:
          // Check for PowerShell in PATH
          try {
-            await execAsync('powershell -Command "Get-Host"', {
-              timeout: 3000
-            })
+            await execAsync('powershell -Command "Get-Host"', { timeout: 3000 })
            return terminal
          } catch {
            try {
@@ -393,9 +384,7 @@ class CodeToolsService {
        const binDir = path.join(os.homedir(), '.cherrystudio', 'bin')
        const executablePath = path.join(binDir, executableName + (isWin ? '.exe' : ''))

-        const { stdout } = await execAsync(`"${executablePath}" --version`, {
-          timeout: 10000
-        })
+        const { stdout } = await execAsync(`"${executablePath}" --version`, { timeout: 10000 })
        // Extract version number from output (format may vary by tool)
        const versionMatch = stdout.trim().match(/\d+\.\d+\.\d+/)
        installedVersion = versionMatch ? versionMatch[0] : stdout.trim().split(' ')[0]
@@ -436,10 +425,7 @@ class CodeToolsService {
        logger.info(`${packageName} latest version: ${latestVersion}`)

        // Cache the result
-        this.versionCache.set(cacheKey, {
-          version: latestVersion!,
-          timestamp: now
-        })
+        this.versionCache.set(cacheKey, { version: latestVersion!, timestamp: now })
        logger.debug(`Cached latest version for ${packageName}`)
      } catch (error) {
        logger.warn(`Failed to get latest version for ${packageName}:`, error as Error)
--- a/src/main/services/KnowledgeService.ts
+++ b/src/main/services/KnowledgeService.ts
@@ -29,7 +29,7 @@ import Reranker from '@main/knowledge/reranker/Reranker'
 import { fileStorage } from '@main/services/FileStorage'
 import { windowService } from '@main/services/WindowService'
 import { getDataPath } from '@main/utils'
-import { getAllFiles, sanitizeFilename } from '@main/utils/file'
+import { getAllFiles } from '@main/utils/file'
 import { TraceMethod } from '@mcp-trace/trace-core'
 import { MB } from '@shared/config/constant'
 import type { LoaderReturn } from '@shared/config/types'
@@ -147,16 +147,11 @@ class KnowledgeService {
    }
  }

-  private getDbPath = (id: string): string => {
-    // 消除网络搜索requestI d中的特殊字符
-    return path.join(this.storageDir, sanitizeFilename(id, '_'))
-  }
-
  /**
   * Delete knowledge base file
   */
  private deleteKnowledgeFile = (id: string): boolean => {
-    const dbPath = this.getDbPath(id)
+    const dbPath = path.join(this.storageDir, id)
    if (fs.existsSync(dbPath)) {
      try {
        fs.rmSync(dbPath, { recursive: true })
@@ -249,8 +244,7 @@ class KnowledgeService {
      dimensions
    })
    try {
-      const dbPath = this.getDbPath(id)
-      const libSqlDb = new LibSqlDb({ path: dbPath })
+      const libSqlDb = new LibSqlDb({ path: path.join(this.storageDir, id) })
      // Save database instance for later closing
      this.dbInstances.set(id, libSqlDb)

--- a/src/main/services/OvmsManager.ts
+++ b/src/main/services/OvmsManager.ts
@@ -1,586 +0,0 @@
-import { exec } from 'node:child_process'
-import { homedir } from 'node:os'
-import { promisify } from 'node:util'
-
-import { loggerService } from '@logger'
-import * as fs from 'fs-extra'
-import * as path from 'path'
-
-const logger = loggerService.withContext('OvmsManager')
-
-const execAsync = promisify(exec)
-
-interface OvmsProcess {
-  pid: number
-  path: string
-  workingDirectory: string
-}
-
-interface ModelConfig {
-  name: string
-  base_path: string
-}
-
-interface OvmsConfig {
-  mediapipe_config_list: ModelConfig[]
-}
-
-class OvmsManager {
-  private ovms: OvmsProcess | null = null
-
-  /**
-   * Recursively terminate a process and all its child processes
-   * @param pid Process ID to terminate
-   * @returns Promise<{ success: boolean; message?: string }>
-   */
-  private async terminalProcess(pid: number): Promise<{ success: boolean; message?: string }> {
-    try {
-      // Check if the process is running
-      const processCheckCommand = `Get-Process -Id ${pid} -ErrorAction SilentlyContinue | Select-Object Id | ConvertTo-Json`
-      const { stdout: processStdout } = await execAsync(`powershell -Command "${processCheckCommand}"`)
-
-      if (!processStdout.trim()) {
-        logger.info(`Process with PID ${pid} is not running`)
-        return { success: true, message: `Process with PID ${pid} is not running` }
-      }
-
-      // Find child processes
-      const childProcessCommand = `Get-WmiObject -Class Win32_Process | Where-Object { $_.ParentProcessId -eq ${pid} } | Select-Object ProcessId | ConvertTo-Json`
-      const { stdout: childStdout } = await execAsync(`powershell -Command "${childProcessCommand}"`)
-
-      // If there are child processes, terminate them first
-      if (childStdout.trim()) {
-        const childProcesses = JSON.parse(childStdout)
-        const childList = Array.isArray(childProcesses) ? childProcesses : [childProcesses]
-
-        logger.info(`Found ${childList.length} child processes for PID ${pid}`)
-
-        // Recursively terminate each child process
-        for (const childProcess of childList) {
-          const childPid = childProcess.ProcessId
-          logger.info(`Terminating child process PID: ${childPid}`)
-          await this.terminalProcess(childPid)
-        }
-      } else {
-        logger.info(`No child processes found for PID ${pid}`)
-      }
-
-      // Finally, terminate the parent process
-      const killCommand = `Stop-Process -Id ${pid} -Force -ErrorAction SilentlyContinue`
-      await execAsync(`powershell -Command "${killCommand}"`)
-      logger.info(`Terminated process with PID: ${pid}`)
-
-      // Wait for the process to disappear with 5-second timeout
-      const timeout = 5000 // 5 seconds
-      const startTime = Date.now()
-
-      while (Date.now() - startTime < timeout) {
-        const checkCommand = `Get-Process -Id ${pid} -ErrorAction SilentlyContinue | Select-Object Id | ConvertTo-Json`
-        const { stdout: checkStdout } = await execAsync(`powershell -Command "${checkCommand}"`)
-
-        if (!checkStdout.trim()) {
-          logger.info(`Process with PID ${pid} has disappeared`)
-          return { success: true, message: `Process ${pid} and all child processes terminated successfully` }
-        }
-
-        // Wait 300ms before checking again
-        await new Promise((resolve) => setTimeout(resolve, 300))
-      }
-
-      logger.warn(`Process with PID ${pid} did not disappear within timeout`)
-      return { success: false, message: `Process ${pid} did not disappear within 5 seconds` }
-    } catch (error) {
-      logger.error(`Failed to terminate process ${pid}:`, error as Error)
-      return { success: false, message: `Failed to terminate process ${pid}` }
-    }
-  }
-
-  /**
-   * Stop OVMS process if it's running
-   * @returns Promise<{ success: boolean; message?: string }>
-   */
-  public async stopOvms(): Promise<{ success: boolean; message?: string }> {
-    try {
-      // Check if OVMS process is running
-      const psCommand = `Get-Process -Name "ovms" -ErrorAction SilentlyContinue | Select-Object Id, Path | ConvertTo-Json`
-      const { stdout } = await execAsync(`powershell -Command "${psCommand}"`)
-
-      if (!stdout.trim()) {
-        logger.info('OVMS process is not running')
-        return { success: true, message: 'OVMS process is not running' }
-      }
-
-      const processes = JSON.parse(stdout)
-      const processList = Array.isArray(processes) ? processes : [processes]
-
-      if (processList.length === 0) {
-        logger.info('OVMS process is not running')
-        return { success: true, message: 'OVMS process is not running' }
-      }
-
-      // Terminate all OVMS processes using terminalProcess
-      for (const process of processList) {
-        const result = await this.terminalProcess(process.Id)
-        if (!result.success) {
-          logger.error(`Failed to terminate OVMS process with PID: ${process.Id}, ${result.message}`)
-          return { success: false, message: `Failed to terminate OVMS process: ${result.message}` }
-        }
-        logger.info(`Terminated OVMS process with PID: ${process.Id}`)
-      }
-
-      // Reset the ovms instance
-      this.ovms = null
-
-      logger.info('OVMS process stopped successfully')
-      return { success: true, message: 'OVMS process stopped successfully' }
-    } catch (error) {
-      logger.error(`Failed to stop OVMS process: ${error}`)
-      return { success: false, message: 'Failed to stop OVMS process' }
-    }
-  }
-
-  /**
-   * Run OVMS by ensuring config.json exists and executing run.bat
-   * @returns Promise<{ success: boolean; message?: string }>
-   */
-  public async runOvms(): Promise<{ success: boolean; message?: string }> {
-    const homeDir = homedir()
-    const ovmsDir = path.join(homeDir, '.cherrystudio', 'ovms', 'ovms')
-    const configPath = path.join(ovmsDir, 'models', 'config.json')
-    const runBatPath = path.join(ovmsDir, 'run.bat')
-
-    try {
-      // Check if config.json exists, if not create it with default content
-      if (!(await fs.pathExists(configPath))) {
-        logger.info(`Config file does not exist, creating: ${configPath}`)
-
-        // Ensure the models directory exists
-        await fs.ensureDir(path.dirname(configPath))
-
-        // Create config.json with default content
-        const defaultConfig = {
-          mediapipe_config_list: [],
-          model_config_list: []
-        }
-
-        await fs.writeJson(configPath, defaultConfig, { spaces: 2 })
-        logger.info(`Config file created: ${configPath}`)
-      }
-
-      // Check if run.bat exists
-      if (!(await fs.pathExists(runBatPath))) {
-        logger.error(`run.bat not found at: ${runBatPath}`)
-        return { success: false, message: 'run.bat not found' }
-      }
-
-      // Run run.bat without waiting for it to complete
-      logger.info(`Starting OVMS with run.bat: ${runBatPath}`)
-      exec(`"${runBatPath}"`, { cwd: ovmsDir }, (error) => {
-        if (error) {
-          logger.error(`Error running run.bat: ${error}`)
-        }
-      })
-
-      logger.info('OVMS started successfully')
-      return { success: true }
-    } catch (error) {
-      logger.error(`Failed to run OVMS: ${error}`)
-      return { success: false, message: 'Failed to run OVMS' }
-    }
-  }
-
-  /**
-   * Get OVMS status - checks installation and running status
-   * @returns 'not-installed' | 'not-running' | 'running'
-   */
-  public async getOvmsStatus(): Promise<'not-installed' | 'not-running' | 'running'> {
-    const homeDir = homedir()
-    const ovmsPath = path.join(homeDir, '.cherrystudio', 'ovms', 'ovms', 'ovms.exe')
-
-    try {
-      // Check if OVMS executable exists
-      if (!(await fs.pathExists(ovmsPath))) {
-        logger.info(`OVMS executable not found at: ${ovmsPath}`)
-        return 'not-installed'
-      }
-
-      // Check if OVMS process is running
-      //const psCommand = `Get-Process -Name "ovms" -ErrorAction SilentlyContinue | Where-Object { $_.Path -eq "${ovmsPath.replace(/\\/g, '\\\\')}" } | Select-Object Id | ConvertTo-Json`;
-      //const { stdout } = await execAsync(`powershell -Command "${psCommand}"`);
-      const psCommand = `Get-Process -Name "ovms" -ErrorAction SilentlyContinue | Select-Object Id, Path | ConvertTo-Json`
-      const { stdout } = await execAsync(`powershell -Command "${psCommand}"`)
-
-      if (!stdout.trim()) {
-        logger.info('OVMS process not running')
-        return 'not-running'
-      }
-
-      const processes = JSON.parse(stdout)
-      const processList = Array.isArray(processes) ? processes : [processes]
-
-      if (processList.length > 0) {
-        logger.info('OVMS process is running')
-        return 'running'
-      } else {
-        logger.info('OVMS process not running')
-        return 'not-running'
-      }
-    } catch (error) {
-      logger.info(`Failed to check OVMS status: ${error}`)
-      return 'not-running'
-    }
-  }
-
-  /**
-   * Initialize OVMS by finding the executable path and working directory
-   */
-  public async initializeOvms(): Promise<boolean> {
-    // Use PowerShell to find ovms.exe processes with their paths
-    const psCommand = `Get-Process -Name "ovms" -ErrorAction SilentlyContinue | Select-Object Id, Path | ConvertTo-Json`
-    const { stdout } = await execAsync(`powershell -Command "${psCommand}"`)
-
-    if (!stdout.trim()) {
-      logger.error('Command to find OVMS process returned no output')
-      return false
-    }
-    logger.debug(`OVMS process output: ${stdout}`)
-
-    const processes = JSON.parse(stdout)
-    const processList = Array.isArray(processes) ? processes : [processes]
-
-    // Find the first process with a valid path
-    for (const process of processList) {
-      this.ovms = {
-        pid: process.Id,
-        path: process.Path,
-        workingDirectory: path.dirname(process.Path)
-      }
-      return true
-    }
-
-    return this.ovms !== null
-  }
-
-  /**
-   * Check if the Model Name and ID are valid, they are valid only if they are not used in the config.json
-   * @param modelName Name of the model to check
-   * @param modelId ID of the model to check
-   */
-  public async isNameAndIDAvalid(modelName: string, modelId: string): Promise<boolean> {
-    if (!modelName || !modelId) {
-      logger.error('Model name and ID cannot be empty')
-      return false
-    }
-
-    const homeDir = homedir()
-    const configPath = path.join(homeDir, '.cherrystudio', 'ovms', 'ovms', 'models', 'config.json')
-    try {
-      if (!(await fs.pathExists(configPath))) {
-        logger.warn(`Config file does not exist: ${configPath}`)
-        return false
-      }
-
-      const config: OvmsConfig = await fs.readJson(configPath)
-      if (!config.mediapipe_config_list) {
-        logger.warn(`No mediapipe_config_list found in config: ${configPath}`)
-        return false
-      }
-
-      // Check if the model name or ID already exists in the config
-      const exists = config.mediapipe_config_list.some(
-        (model) => model.name === modelName || model.base_path === modelId
-      )
-      if (exists) {
-        logger.warn(`Model with name "${modelName}" or ID "${modelId}" already exists in the config`)
-        return false
-      }
-    } catch (error) {
-      logger.error(`Failed to check model existence: ${error}`)
-      return false
-    }
-
-    return true
-  }
-
-  private async applyModelPath(modelDirPath: string): Promise<boolean> {
-    const homeDir = homedir()
-    const patchDir = path.join(homeDir, '.cherrystudio', 'ovms', 'patch')
-    if (!(await fs.pathExists(patchDir))) {
-      return true
-    }
-
-    const modelId = path.basename(modelDirPath)
-
-    // get all sub directories in patchDir
-    const patchs = await fs.readdir(patchDir)
-    for (const patch of patchs) {
-      const fullPatchPath = path.join(patchDir, patch)
-
-      if (fs.lstatSync(fullPatchPath).isDirectory()) {
-        if (modelId.toLowerCase().includes(patch.toLowerCase())) {
-          // copy all files from fullPath to modelDirPath
-          try {
-            const files = await fs.readdir(fullPatchPath)
-            for (const file of files) {
-              const srcFile = path.join(fullPatchPath, file)
-              const destFile = path.join(modelDirPath, file)
-              await fs.copyFile(srcFile, destFile)
-            }
-          } catch (error) {
-            logger.error(`Failed to copy files from ${fullPatchPath} to ${modelDirPath}: ${error}`)
-            return false
-          }
-          logger.info(`Applied patchs for model ${modelId}`)
-          return true
-        }
-      }
-    }
-
-    return true
-  }
-
-  /**
-   * Add a model to OVMS by downloading it
-   * @param modelName Name of the model to add
-   * @param modelId ID of the model to download
-   * @param modelSource Model Source: huggingface, hf-mirror and modelscope, default is huggingface
-   * @param task Task type: text_generation, embedding, rerank, image_generation
-   */
-  public async addModel(
-    modelName: string,
-    modelId: string,
-    modelSource: string,
-    task: string = 'text_generation'
-  ): Promise<{ success: boolean; message?: string }> {
-    logger.info(`Adding model: ${modelName} with ID: ${modelId}, Source: ${modelSource}, Task: ${task}`)
-
-    const homeDir = homedir()
-    const ovdndDir = path.join(homeDir, '.cherrystudio', 'ovms', 'ovms')
-    const pathModel = path.join(ovdndDir, 'models', modelId)
-
-    try {
-      // check the ovdnDir+'models'+modelId exist or not
-      if (await fs.pathExists(pathModel)) {
-        logger.error(`Model with ID ${modelId} already exists`)
-        return { success: false, message: 'Model ID already exists!' }
-      }
-
-      // remove the model directory if it exists
-      if (await fs.pathExists(pathModel)) {
-        logger.info(`Removing existing model directory: ${pathModel}`)
-        await fs.remove(pathModel)
-      }
-
-      // Use ovdnd.exe for downloading instead of ovms.exe
-      const ovdndPath = path.join(ovdndDir, 'ovdnd.exe')
-      const command =
-        `"${ovdndPath}" --pull ` +
-        `--model_repository_path "${ovdndDir}/models" ` +
-        `--source_model "${modelId}" ` +
-        `--model_name "${modelName}" ` +
-        `--target_device GPU ` +
-        `--task ${task} ` +
-        `--overwrite_models`
-
-      const env: Record<string, string | undefined> = {
-        ...process.env,
-        OVMS_DIR: ovdndDir,
-        PYTHONHOME: path.join(ovdndDir, 'python'),
-        PATH: `${process.env.PATH};${ovdndDir};${path.join(ovdndDir, 'python')}`
-      }
-
-      if (modelSource) {
-        env.HF_ENDPOINT = modelSource
-      }
-
-      logger.info(`Running command: ${command} from ${modelSource}`)
-      const { stdout } = await execAsync(command, { env: env, cwd: ovdndDir })
-
-      logger.info('Model download completed')
-      logger.debug(`Command output: ${stdout}`)
-    } catch (error) {
-      // remove ovdnDir+'models'+modelId if it exists
-      if (await fs.pathExists(pathModel)) {
-        logger.info(`Removing failed model directory: ${pathModel}`)
-        await fs.remove(pathModel)
-      }
-      logger.error(`Failed to add model: ${error}`)
-      return {
-        success: false,
-        message: `Download model ${modelId} failed, please check following items and try it again:<p>- the model id</p><p>- network connection and proxy</p>`
-      }
-    }
-
-    // Update config file
-    if (!(await this.updateModelConfig(modelName, modelId))) {
-      logger.error('Failed to update model config')
-      return { success: false, message: 'Failed to update model config' }
-    }
-
-    if (!(await this.applyModelPath(pathModel))) {
-      logger.error('Failed to apply model patchs')
-      return { success: false, message: 'Failed to apply model patchs' }
-    }
-
-    logger.info(`Model ${modelName} added successfully with ID ${modelId}`)
-    return { success: true }
-  }
-
-  /**
-   * Stop the model download process if it's running
-   * @returns Promise<{ success: boolean; message?: string }>
-   */
-  public async stopAddModel(): Promise<{ success: boolean; message?: string }> {
-    try {
-      // Check if ovdnd.exe process is running
-      const psCommand = `Get-Process -Name "ovdnd" -ErrorAction SilentlyContinue | Select-Object Id, Path | ConvertTo-Json`
-      const { stdout } = await execAsync(`powershell -Command "${psCommand}"`)
-
-      if (!stdout.trim()) {
-        logger.info('ovdnd process is not running')
-        return { success: true, message: 'Model download process is not running' }
-      }
-
-      const processes = JSON.parse(stdout)
-      const processList = Array.isArray(processes) ? processes : [processes]
-
-      if (processList.length === 0) {
-        logger.info('ovdnd process is not running')
-        return { success: true, message: 'Model download process is not running' }
-      }
-
-      // Terminate all ovdnd processes
-      for (const process of processList) {
-        this.terminalProcess(process.Id)
-      }
-
-      logger.info('Model download process stopped successfully')
-      return { success: true, message: 'Model download process stopped successfully' }
-    } catch (error) {
-      logger.error(`Failed to stop model download process: ${error}`)
-      return { success: false, message: 'Failed to stop model download process' }
-    }
-  }
-
-  /**
-   * check if the model id exists in the OVMS configuration
-   * @param modelId ID of the model to check
-   */
-  public async checkModelExists(modelId: string): Promise<boolean> {
-    const homeDir = homedir()
-    const ovmsDir = path.join(homeDir, '.cherrystudio', 'ovms', 'ovms')
-    const configPath = path.join(ovmsDir, 'models', 'config.json')
-
-    try {
-      if (!(await fs.pathExists(configPath))) {
-        logger.warn(`Config file does not exist: ${configPath}`)
-        return false
-      }
-
-      const config: OvmsConfig = await fs.readJson(configPath)
-      if (!config.mediapipe_config_list) {
-        logger.warn('No mediapipe_config_list found in config')
-        return false
-      }
-
-      return config.mediapipe_config_list.some((model) => model.base_path === modelId)
-    } catch (error) {
-      logger.error(`Failed to check model existence: ${error}`)
-      return false
-    }
-  }
-
-  /**
-   * Update the model configuration file
-   */
-  public async updateModelConfig(modelName: string, modelId: string): Promise<boolean> {
-    const homeDir = homedir()
-    const ovmsDir = path.join(homeDir, '.cherrystudio', 'ovms', 'ovms')
-    const configPath = path.join(ovmsDir, 'models', 'config.json')
-
-    try {
-      // Ensure the models directory exists
-      await fs.ensureDir(path.dirname(configPath))
-      let config: OvmsConfig
-
-      // Read existing config or create new one
-      if (await fs.pathExists(configPath)) {
-        config = await fs.readJson(configPath)
-      } else {
-        config = { mediapipe_config_list: [] }
-      }
-
-      // Ensure mediapipe_config_list exists
-      if (!config.mediapipe_config_list) {
-        config.mediapipe_config_list = []
-      }
-
-      // Add new model config
-      const newModelConfig: ModelConfig = {
-        name: modelName,
-        base_path: modelId
-      }
-
-      // Check if model already exists, if so, update it
-      const existingIndex = config.mediapipe_config_list.findIndex((model) => model.base_path === modelId)
-
-      if (existingIndex >= 0) {
-        config.mediapipe_config_list[existingIndex] = newModelConfig
-        logger.info(`Updated existing model config: ${modelName}`)
-      } else {
-        config.mediapipe_config_list.push(newModelConfig)
-        logger.info(`Added new model config: ${modelName}`)
-      }
-
-      // Write config back to file
-      await fs.writeJson(configPath, config, { spaces: 2 })
-      logger.info(`Config file updated: ${configPath}`)
-    } catch (error) {
-      logger.error(`Failed to update model config: ${error}`)
-      return false
-    }
-    return true
-  }
-
-  /**
-   * Get all models from OVMS config, filtered for image generation models
-   * @returns Array of model configurations
-   */
-  public async getModels(): Promise<ModelConfig[]> {
-    const homeDir = homedir()
-    const ovmsDir = path.join(homeDir, '.cherrystudio', 'ovms', 'ovms')
-    const configPath = path.join(ovmsDir, 'models', 'config.json')
-
-    try {
-      if (!(await fs.pathExists(configPath))) {
-        logger.warn(`Config file does not exist: ${configPath}`)
-        return []
-      }
-
-      const config: OvmsConfig = await fs.readJson(configPath)
-      if (!config.mediapipe_config_list) {
-        logger.warn('No mediapipe_config_list found in config')
-        return []
-      }
-
-      // Filter models for image generation (SD, Stable-Diffusion, Stable Diffusion, FLUX)
-      const imageGenerationModels = config.mediapipe_config_list.filter((model) => {
-        const modelName = model.name.toLowerCase()
-        return (
-          modelName.startsWith('sd') ||
-          modelName.startsWith('stable-diffusion') ||
-          modelName.startsWith('stable diffusion') ||
-          modelName.startsWith('flux')
-        )
-      })
-
-      logger.info(`Found ${imageGenerationModels.length} image generation models`)
-      return imageGenerationModels
-    } catch (error) {
-      logger.error(`Failed to get models: ${error}`)
-      return []
-    }
-  }
-}
-
-export default OvmsManager
--- a/src/main/services/tests/AppUpdater.test.ts
+++ b/src/main/services/tests/AppUpdater.test.ts
@@ -274,4 +274,46 @@ describe('AppUpdater', () => {
      expect(result.releaseNotes).toBeNull()
    })
  })
+
+  describe('formatReleaseNotes', () => {
+    it('should format string release notes with markers', () => {
+      vi.mocked(configManager.getLanguage).mockReturnValue('en-US')
+      const notes = `<!--LANG:en-->English<!--LANG:zh-CN-->中文<!--LANG:END-->`
+
+      const result = (appUpdater as any).formatReleaseNotes(notes)
+
+      expect(result).toBe('English')
+    })
+
+    it('should format string release notes without markers', () => {
+      const notes = 'Simple notes'
+
+      const result = (appUpdater as any).formatReleaseNotes(notes)
+
+      expect(result).toBe('Simple notes')
+    })
+
+    it('should format array release notes', () => {
+      const notes = [
+        { version: '1.0.0', note: 'Note 1' },
+        { version: '1.0.1', note: 'Note 2' }
+      ]
+
+      const result = (appUpdater as any).formatReleaseNotes(notes)
+
+      expect(result).toBe('Note 1\nNote 2')
+    })
+
+    it('should handle null release notes', () => {
+      const result = (appUpdater as any).formatReleaseNotes(null)
+
+      expect(result).toBe('')
+    })
+
+    it('should handle undefined release notes', () => {
+      const result = (appUpdater as any).formatReleaseNotes(undefined)
+
+      expect(result).toBe('')
+    })
+  })
 })
--- a/src/main/utils/file.ts
+++ b/src/main/utils/file.ts
@@ -1,4 +1,3 @@
-import { createHash } from 'node:crypto'
 import * as fs from 'node:fs'
 import { readFile } from 'node:fs/promises'
 import os from 'node:os'
@@ -265,12 +264,11 @@ export async function scanDir(dirPath: string, depth = 0, basePath?: string): Pr

    if (entry.isDirectory() && options.includeDirectories) {
      const stats = await fs.promises.stat(entryPath)
-      const externalDirPath = entryPath.replace(/\\/g, '/')
      const dirTreeNode: NotesTreeNode = {
-        id: createHash('sha1').update(externalDirPath).digest('hex'),
+        id: uuidv4(),
        name: entry.name,
        treePath: treePath,
-        externalPath: externalDirPath,
+        externalPath: entryPath,
        createdAt: stats.birthtime.toISOString(),
        updatedAt: stats.mtime.toISOString(),
        type: 'folder',
@@ -301,12 +299,11 @@ export async function scanDir(dirPath: string, depth = 0, basePath?: string): Pr
        ? `/${dirRelativePath.replace(/\\/g, '/')}/${nameWithoutExt}`
        : `/${nameWithoutExt}`

-      const externalFilePath = entryPath.replace(/\\/g, '/')
      const fileTreeNode: NotesTreeNode = {
-        id: createHash('sha1').update(externalFilePath).digest('hex'),
+        id: uuidv4(),
        name: name,
        treePath: fileTreePath,
-        externalPath: externalFilePath,
+        externalPath: entryPath,
        createdAt: stats.birthtime.toISOString(),
        updatedAt: stats.mtime.toISOString(),
        type: 'file'
--- a/src/preload/index.ts
+++ b/src/preload/index.ts
@@ -51,7 +51,7 @@ const api = {
  setProxy: (proxy: string | undefined, bypassRules?: string) =>
    ipcRenderer.invoke(IpcChannel.App_Proxy, proxy, bypassRules),
  checkForUpdate: () => ipcRenderer.invoke(IpcChannel.App_CheckForUpdate),
-  quitAndInstall: () => ipcRenderer.invoke(IpcChannel.App_QuitAndInstall),
+  showUpdateDialog: () => ipcRenderer.invoke(IpcChannel.App_ShowUpdateDialog),
  setLanguage: (lang: string) => ipcRenderer.invoke(IpcChannel.App_SetLanguage, lang),
  setEnableSpellCheck: (isEnable: boolean) => ipcRenderer.invoke(IpcChannel.App_SetEnableSpellCheck, isEnable),
  setSpellCheckLanguages: (languages: string[]) => ipcRenderer.invoke(IpcChannel.App_SetSpellCheckLanguages, languages),
@@ -95,8 +95,7 @@ const api = {
  },
  system: {
    getDeviceType: () => ipcRenderer.invoke(IpcChannel.System_GetDeviceType),
-    getHostname: () => ipcRenderer.invoke(IpcChannel.System_GetHostname),
-    getCpuName: () => ipcRenderer.invoke(IpcChannel.System_GetCpuName)
+    getHostname: () => ipcRenderer.invoke(IpcChannel.System_GetHostname)
  },
  devTools: {
    toggle: () => ipcRenderer.invoke(IpcChannel.System_ToggleDevTools)
@@ -221,7 +220,7 @@ const api = {
    create: (base: KnowledgeBaseParams, context?: SpanContext) =>
      tracedInvoke(IpcChannel.KnowledgeBase_Create, context, base),
    reset: (base: KnowledgeBaseParams) => ipcRenderer.invoke(IpcChannel.KnowledgeBase_Reset, base),
-    delete: (id: string) => ipcRenderer.invoke(IpcChannel.KnowledgeBase_Delete, id),
+    delete: (base: KnowledgeBaseParams, id: string) => ipcRenderer.invoke(IpcChannel.KnowledgeBase_Delete, base, id),
    add: ({
      base,
      item,
@@ -286,16 +285,6 @@ const api = {
    clearAuthCache: (projectId: string, clientEmail?: string) =>
      ipcRenderer.invoke(IpcChannel.VertexAI_ClearAuthCache, projectId, clientEmail)
  },
-  ovms: {
-    addModel: (modelName: string, modelId: string, modelSource: string, task: string) =>
-      ipcRenderer.invoke(IpcChannel.Ovms_AddModel, modelName, modelId, modelSource, task),
-    stopAddModel: () => ipcRenderer.invoke(IpcChannel.Ovms_StopAddModel),
-    getModels: () => ipcRenderer.invoke(IpcChannel.Ovms_GetModels),
-    isRunning: () => ipcRenderer.invoke(IpcChannel.Ovms_IsRunning),
-    getStatus: () => ipcRenderer.invoke(IpcChannel.Ovms_GetStatus),
-    runOvms: () => ipcRenderer.invoke(IpcChannel.Ovms_RunOVMS),
-    stopOvms: () => ipcRenderer.invoke(IpcChannel.Ovms_StopOVMS)
-  },
  config: {
    set: (key: string, value: any, isNotify: boolean = false) =>
      ipcRenderer.invoke(IpcChannel.Config_Set, key, value, isNotify),
@@ -361,7 +350,6 @@ const api = {
  getBinaryPath: (name: string) => ipcRenderer.invoke(IpcChannel.App_GetBinaryPath, name),
  installUVBinary: () => ipcRenderer.invoke(IpcChannel.App_InstallUvBinary),
  installBunBinary: () => ipcRenderer.invoke(IpcChannel.App_InstallBunBinary),
-  installOvmsBinary: () => ipcRenderer.invoke(IpcChannel.App_InstallOvmsBinary),
  protocol: {
    onReceiveData: (callback: (data: { url: string; params: any }) => void) => {
      const listener = (_event: Electron.IpcRendererEvent, data: { url: string; params: any }) => {
--- a/src/renderer/src/aiCore/chunk/AiSdkToChunkAdapter.ts
+++ b/src/renderer/src/aiCore/chunk/AiSdkToChunkAdapter.ts
@@ -22,8 +22,6 @@ export class AiSdkToChunkAdapter {
  private accumulate: boolean | undefined
  private isFirstChunk = true
  private enableWebSearch: boolean = false
-  private responseStartTimestamp: number | null = null
-  private firstTokenTimestamp: number | null = null

  constructor(
    private onChunk: (chunk: Chunk) => void,
@@ -36,17 +34,6 @@ export class AiSdkToChunkAdapter {
    this.enableWebSearch = enableWebSearch || false
  }

-  private markFirstTokenIfNeeded() {
-    if (this.firstTokenTimestamp === null && this.responseStartTimestamp !== null) {
-      this.firstTokenTimestamp = Date.now()
-    }
-  }
-
-  private resetTimingState() {
-    this.responseStartTimestamp = null
-    this.firstTokenTimestamp = null
-  }
-
  /**
   * 处理 AI SDK 流结果
   * @param aiSdkResult AI SDK 的流结果对象
@@ -74,8 +61,6 @@ export class AiSdkToChunkAdapter {
      webSearchResults: [],
      reasoningId: ''
    }
-    this.resetTimingState()
-    this.responseStartTimestamp = Date.now()
    // Reset link converter state at the start of stream
    this.isFirstChunk = true

@@ -88,7 +73,6 @@ export class AiSdkToChunkAdapter {
          if (this.enableWebSearch) {
            const remainingText = flushLinkConverterBuffer()
            if (remainingText) {
-              this.markFirstTokenIfNeeded()
              this.onChunk({
                type: ChunkType.TEXT_DELTA,
                text: remainingText
@@ -103,7 +87,6 @@ export class AiSdkToChunkAdapter {
      }
    } finally {
      reader.releaseLock()
-      this.resetTimingState()
    }
  }

@@ -154,7 +137,6 @@ export class AiSdkToChunkAdapter {

        // Only emit chunk if there's text to send
        if (finalText) {
-          this.markFirstTokenIfNeeded()
          this.onChunk({
            type: ChunkType.TEXT_DELTA,
            text: this.accumulate ? final.text : finalText
@@ -179,18 +161,16 @@ export class AiSdkToChunkAdapter {
        break
      case 'reasoning-delta':
        final.reasoningContent += chunk.text || ''
-        if (chunk.text) {
-          this.markFirstTokenIfNeeded()
-        }
        this.onChunk({
          type: ChunkType.THINKING_DELTA,
-          text: final.reasoningContent || ''
+          text: final.reasoningContent || '',
+          thinking_millsec: (chunk.providerMetadata?.metadata?.thinking_millsec as number) || 0
        })
        break
      case 'reasoning-end':
        this.onChunk({
          type: ChunkType.THINKING_COMPLETE,
-          text: final.reasoningContent || ''
+          text: (chunk.providerMetadata?.metadata?.thinking_content as string) || final.reasoningContent
        })
        final.reasoningContent = ''
        break
@@ -281,37 +261,44 @@ export class AiSdkToChunkAdapter {
        break
      }

-      case 'finish': {
-        const usage = {
-          completion_tokens: chunk.totalUsage?.outputTokens || 0,
-          prompt_tokens: chunk.totalUsage?.inputTokens || 0,
-          total_tokens: chunk.totalUsage?.totalTokens || 0
-        }
-        const metrics = this.buildMetrics(chunk.totalUsage)
-        const baseResponse = {
-          text: final.text || '',
-          reasoning_content: final.reasoningContent || ''
-        }
-
+      case 'finish':
        this.onChunk({
          type: ChunkType.BLOCK_COMPLETE,
          response: {
-            ...baseResponse,
-            usage: { ...usage },
-            metrics: metrics ? { ...metrics } : undefined
+            text: final.text || '',
+            reasoning_content: final.reasoningContent || '',
+            usage: {
+              completion_tokens: chunk.totalUsage.outputTokens || 0,
+              prompt_tokens: chunk.totalUsage.inputTokens || 0,
+              total_tokens: chunk.totalUsage.totalTokens || 0
+            },
+            metrics: chunk.totalUsage
+              ? {
+                  completion_tokens: chunk.totalUsage.outputTokens || 0,
+                  time_completion_millsec: 0
+                }
+              : undefined
          }
        })
        this.onChunk({
          type: ChunkType.LLM_RESPONSE_COMPLETE,
          response: {
-            ...baseResponse,
-            usage: { ...usage },
-            metrics: metrics ? { ...metrics } : undefined
+            text: final.text || '',
+            reasoning_content: final.reasoningContent || '',
+            usage: {
+              completion_tokens: chunk.totalUsage.outputTokens || 0,
+              prompt_tokens: chunk.totalUsage.inputTokens || 0,
+              total_tokens: chunk.totalUsage.totalTokens || 0
+            },
+            metrics: chunk.totalUsage
+              ? {
+                  completion_tokens: chunk.totalUsage.outputTokens || 0,
+                  time_completion_millsec: 0
+                }
+              : undefined
          }
        })
-        this.resetTimingState()
        break
-      }

      // === 源和文件相关事件 ===
      case 'source':
@@ -347,34 +334,6 @@ export class AiSdkToChunkAdapter {
      default:
    }
  }
-
-  private buildMetrics(totalUsage?: {
-    inputTokens?: number | null
-    outputTokens?: number | null
-    totalTokens?: number | null
-  }) {
-    if (!totalUsage) {
-      return undefined
-    }
-
-    const completionTokens = totalUsage.outputTokens ?? 0
-    const now = Date.now()
-    const start = this.responseStartTimestamp ?? now
-    const firstToken = this.firstTokenTimestamp
-    const timeFirstToken = Math.max(firstToken != null ? firstToken - start : 0, 0)
-    const baseForCompletion = firstToken ?? start
-    let timeCompletion = Math.max(now - baseForCompletion, 0)
-
-    if (timeCompletion === 0 && completionTokens > 0) {
-      timeCompletion = 1
-    }
-
-    return {
-      completion_tokens: completionTokens,
-      time_first_token_millsec: timeFirstToken,
-      time_completion_millsec: timeCompletion
-    }
-  }
 }

 export default AiSdkToChunkAdapter
--- a/src/renderer/src/aiCore/legacy/clients/ApiClientFactory.ts
+++ b/src/renderer/src/aiCore/legacy/clients/ApiClientFactory.ts
@@ -12,7 +12,6 @@ import { VertexAPIClient } from './gemini/VertexAPIClient'
 import { NewAPIClient } from './newapi/NewAPIClient'
 import { OpenAIAPIClient } from './openai/OpenAIApiClient'
 import { OpenAIResponseAPIClient } from './openai/OpenAIResponseAPIClient'
-import { OVMSClient } from './ovms/OVMSClient'
 import { PPIOAPIClient } from './ppio/PPIOAPIClient'
 import { ZhipuAPIClient } from './zhipu/ZhipuAPIClient'

@@ -64,12 +63,6 @@ export class ApiClientFactory {
      return instance
    }

-    if (provider.id === 'ovms') {
-      logger.debug(`Creating OVMSClient for provider: ${provider.id}`)
-      instance = new OVMSClient(provider) as BaseApiClient
-      return instance
-    }
-
    // 然后检查标准的 Provider Type
    switch (provider.type) {
      case 'openai':
--- a/src/renderer/src/aiCore/legacy/clients/ovms/OVMSClient.ts
+++ b/src/renderer/src/aiCore/legacy/clients/ovms/OVMSClient.ts
@@ -1,56 +0,0 @@
-import { loggerService } from '@logger'
-import { isSupportedModel } from '@renderer/config/models'
-import { objectKeys, Provider } from '@renderer/types'
-import OpenAI from 'openai'
-
-import { OpenAIAPIClient } from '../openai/OpenAIApiClient'
-
-const logger = loggerService.withContext('OVMSClient')
-
-export class OVMSClient extends OpenAIAPIClient {
-  constructor(provider: Provider) {
-    super(provider)
-  }
-
-  override async listModels(): Promise<OpenAI.Models.Model[]> {
-    try {
-      const sdk = await this.getSdkInstance()
-
-      const chatModelsResponse = await sdk.request({
-        method: 'get',
-        path: '../v1/config'
-      })
-      logger.debug(`Chat models response: ${JSON.stringify(chatModelsResponse)}`)
-
-      // Parse the config response to extract model information
-      const config = chatModelsResponse as Record<string, any>
-      const models = objectKeys(config)
-        .map((modelName) => {
-          const modelInfo = config[modelName]
-
-          // Check if model has at least one version with "AVAILABLE" state
-          const hasAvailableVersion = modelInfo?.model_version_status?.some(
-            (versionStatus: any) => versionStatus?.state === 'AVAILABLE'
-          )
-
-          if (hasAvailableVersion) {
-            return {
-              id: modelName,
-              object: 'model' as const,
-              owned_by: 'ovms',
-              created: Date.now()
-            }
-          }
-          return null // Skip models without available versions
-        })
-        .filter(Boolean) // Remove null entries
-      logger.debug(`Processed models: ${JSON.stringify(models)}`)
-
-      // Filter out unsupported models
-      return models.filter((model): model is OpenAI.Models.Model => model !== null && isSupportedModel(model))
-    } catch (error) {
-      logger.error(`Error listing OVMS models: ${error}`)
-      return []
-    }
-  }
-}
--- a/src/renderer/src/aiCore/plugins/PluginBuilder.ts
+++ b/src/renderer/src/aiCore/plugins/PluginBuilder.ts
@@ -5,6 +5,7 @@ import { getEnableDeveloperMode } from '@renderer/hooks/useSettings'
 import { Assistant } from '@renderer/types'

 import { AiSdkMiddlewareConfig } from '../middleware/AiSdkMiddlewareBuilder'
+import reasoningTimePlugin from './reasoningTimePlugin'
 import { searchOrchestrationPlugin } from './searchOrchestrationPlugin'
 import { createTelemetryPlugin } from './telemetryPlugin'

@@ -38,9 +39,9 @@ export function buildPlugins(
  }

  // 3. 推理模型时添加推理插件
-  // if (middlewareConfig.enableReasoning) {
-  //   plugins.push(reasoningTimePlugin)
-  // }
+  if (middlewareConfig.enableReasoning) {
+    plugins.push(reasoningTimePlugin)
+  }

  // 4. 启用Prompt工具调用时添加工具插件
  if (middlewareConfig.isPromptToolUse) {
--- a/src/renderer/src/aiCore/prepareParams/parameterBuilder.ts
+++ b/src/renderer/src/aiCore/prepareParams/parameterBuilder.ts
@@ -23,7 +23,6 @@ import { CherryWebSearchConfig } from '@renderer/store/websearch'
 import { type Assistant, type MCPTool, type Provider } from '@renderer/types'
 import type { StreamTextParams } from '@renderer/types/aiCoreTypes'
 import { mapRegexToPatterns } from '@renderer/utils/blacklistMatchPattern'
-import { replacePromptVariables } from '@renderer/utils/prompt'
 import type { ModelMessage, Tool } from 'ai'
 import { stepCountIs } from 'ai'

@@ -167,7 +166,7 @@ export async function buildStreamTextParams(
    params.tools = tools
  }
  if (assistant.prompt) {
-    params.system = await replacePromptVariables(assistant.prompt, model.name)
+    params.system = assistant.prompt
  }
  logger.debug('params', params)
  return {
--- a/src/renderer/src/aiCore/provider/providerConfig.ts
+++ b/src/renderer/src/aiCore/provider/providerConfig.ts
@@ -18,7 +18,7 @@ import { loggerService } from '@renderer/services/LoggerService'
 import store from '@renderer/store'
 import { isSystemProvider, type Model, type Provider } from '@renderer/types'
 import { formatApiHost } from '@renderer/utils/api'
-import { cloneDeep, trim } from 'lodash'
+import { cloneDeep, isEmpty } from 'lodash'

 import { aihubmixProviderCreator, newApiResolverCreator, vertexAnthropicProviderCreator } from './config'
 import { getAiSdkProviderId } from './factory'
@@ -120,7 +120,7 @@ export function providerToAiSdkConfig(

  // 构建基础配置
  const baseConfig = {
-    baseURL: trim(actualProvider.apiHost),
+    baseURL: actualProvider.apiHost,
    apiKey: getRotatedApiKey(actualProvider)
  }
  // 处理OpenAI模式
@@ -195,10 +195,7 @@ export function providerToAiSdkConfig(
    } else if (baseConfig.baseURL.endsWith('/v1')) {
      baseConfig.baseURL = baseConfig.baseURL.slice(0, -3)
    }
-
-    if (baseConfig.baseURL && !baseConfig.baseURL.includes('publishers/google')) {
-      baseConfig.baseURL = `${baseConfig.baseURL}/v1/projects/${project}/locations/${location}/publishers/google`
-    }
+    baseConfig.baseURL = isEmpty(baseConfig.baseURL) ? '' : baseConfig.baseURL
  }

  // 如果AI SDK支持该provider，使用原生配置
--- a/src/renderer/src/aiCore/tools/KnowledgeSearchTool.ts
+++ b/src/renderer/src/aiCore/tools/KnowledgeSearchTool.ts
@@ -18,13 +18,12 @@ export const knowledgeSearchTool = (
 ) => {
  return tool({
    name: 'builtin_knowledge_search',
-    description: `Knowledge base search tool for retrieving information from user's private knowledge base. This searches your local collection of documents, web content, notes, and other materials you have stored.
+    description: `Search the knowledge base for relevant information using pre-analyzed search intent.

-This tool has been configured with search parameters based on the conversation context:
- Prepared queries: ${extractedKeywords.question.map((q) => `"${q}"`).join(', ')}
- Query rewrite: "${extractedKeywords.rewrite}"
+Pre-extracted search queries: "${extractedKeywords.question.join(', ')}"
+Rewritten query: "${extractedKeywords.rewrite}"

-You can use this tool as-is, or provide additionalContext to refine the search focus within the knowledge base.`,
+Call this tool to execute the search. You can optionally provide additional context to refine the search.`,

    inputSchema: z.object({
      additionalContext: z
--- a/src/renderer/src/aiCore/tools/WebSearchTool.ts
+++ b/src/renderer/src/aiCore/tools/WebSearchTool.ts
@@ -21,17 +21,16 @@ export const webSearchToolWithPreExtractedKeywords = (

  return tool({
    name: 'builtin_web_search',
-    description: `Web search tool for finding current information, news, and real-time data from the internet.
+    description: `Search the web and return citable sources using pre-analyzed search intent.

-This tool has been configured with search parameters based on the conversation context:
- Prepared queries: ${extractedKeywords.question.map((q) => `"${q}"`).join(', ')}${
-      extractedKeywords.links?.length
+Pre-extracted search keywords: "${extractedKeywords.question.join(', ')}"${
+      extractedKeywords.links
        ? `
- Relevant URLs: ${extractedKeywords.links.join(', ')}`
+Relevant links: ${extractedKeywords.links.join(', ')}`
        : ''
    }

-You can use this tool as-is to search with the prepared queries, or provide additionalContext to refine or replace the search terms.`,
+Call this tool to execute the search. You can optionally provide additional context to refine the search.`,

    inputSchema: z.object({
      additionalContext: z
--- a/src/renderer/src/assets/images/apps/3mintop.png
+++ b/src/renderer/src/assets/images/apps/3mintop.png
--- a/src/renderer/src/assets/images/apps/stepfun.png
+++ b/src/renderer/src/assets/images/apps/stepfun.png
--- a/src/renderer/src/assets/images/apps/wpslingxi.webp
+++ b/src/renderer/src/assets/images/apps/wpslingxi.webp
--- a/src/renderer/src/assets/images/apps/yuewen.png
+++ b/src/renderer/src/assets/images/apps/yuewen.png
--- a/src/renderer/src/assets/images/models/gpt-5-codex.png
+++ b/src/renderer/src/assets/images/models/gpt-5-codex.png
--- a/src/renderer/src/assets/images/providers/intel.png
+++ b/src/renderer/src/assets/images/providers/intel.png
--- a/src/renderer/src/assets/images/providers/longcat.png
+++ b/src/renderer/src/assets/images/providers/longcat.png
--- a/src/renderer/src/assets/styles/tailwind.css
+++ b/src/renderer/src/assets/styles/tailwind.css
@@ -163,6 +163,13 @@
  }
 }

+@layer components {
+  @keyframes fadeInWithBlur {
+    from { opacity: 0; filter: blur(2px); }
+     to { opacity: 1; filter: blur(0px); }
+   }
+}
+
 :root {
  background-color: unset;
 }
--- a/src/renderer/src/components/CodeEditor/index.tsx
+++ b/src/renderer/src/components/CodeEditor/index.tsx
@@ -75,15 +75,10 @@ export interface CodeEditorProps {
  /** CSS class name appended to the default `code-editor` class. */
  className?: string
  /**
-   * Whether the editor view is editable.
+   * Whether the editor is editable.
   * @default true
   */
  editable?: boolean
-  /**
-   * Set the editor state to read only but keep some user interactions, e.g., keymaps.
-   * @default false
-   */
-  readOnly?: boolean
  /**
   * Whether the editor is expanded.
   * If true, the height and maxHeight props are ignored.
@@ -119,7 +114,6 @@ const CodeEditor = ({
  style,
  className,
  editable = true,
-  readOnly = false,
  expanded = true,
  wrapped = true
 }: CodeEditorProps) => {
@@ -195,7 +189,6 @@ const CodeEditor = ({
      maxHeight={expanded ? undefined : maxHeight}
      minHeight={minHeight}
      editable={editable}
-      readOnly={readOnly}
      // @ts-ignore 强制使用，见 react-codemirror 的 Example.tsx
      theme={activeCmTheme}
      extensions={customExtensions}
--- a/src/renderer/src/components/HorizontalScrollContainer/index.tsx
+++ b/src/renderer/src/components/HorizontalScrollContainer/index.tsx
@@ -1,4 +1,3 @@
-import { cn } from '@heroui/react'
 import Scrollbar from '@renderer/components/Scrollbar'
 import { ChevronRight } from 'lucide-react'
 import { useEffect, useRef, useState } from 'react'
@@ -18,10 +17,6 @@ export interface HorizontalScrollContainerProps {
  dependencies?: readonly unknown[]
  scrollDistance?: number
  className?: string
-  classNames?: {
-    container?: string
-    content?: string
-  }
  gap?: string
  expandable?: boolean
 }
@@ -31,7 +26,6 @@ const HorizontalScrollContainer: React.FC<HorizontalScrollContainerProps> = ({
  dependencies = [],
  scrollDistance = 200,
  className,
-  classNames,
  gap = '8px',
  expandable = false
 }) => {
@@ -101,16 +95,11 @@ const HorizontalScrollContainer: React.FC<HorizontalScrollContainerProps> = ({

  return (
    <Container
-      className={cn(className, classNames?.container)}
+      className={className}
      $expandable={expandable}
      $disableHoverButton={isScrolledToEnd}
      onClick={expandable ? handleContainerClick : undefined}>
-      <ScrollContent
-        ref={scrollRef}
-        $gap={gap}
-        $isExpanded={isExpanded}
-        $expandable={expandable}
-        className={cn(classNames?.content)}>
+      <ScrollContent ref={scrollRef} $gap={gap} $isExpanded={isExpanded} $expandable={expandable}>
        {children}
      </ScrollContent>
      {canScroll && !isExpanded && !isScrolledToEnd && (
--- a/src/renderer/src/components/MinApp/MinappPopupContainer.tsx
+++ b/src/renderer/src/components/MinApp/MinappPopupContainer.tsx
@@ -549,7 +549,7 @@ const MinappPopupContainer: React.FC = () => {
      {/* 在所有小程序中显示GoogleLoginTip */}
      <GoogleLoginTip isReady={isReady} currentUrl={currentUrl} currentAppId={currentMinappId} />
      {!isReady && (
-        <EmptyView style={{ backgroundColor: 'var(--color-background-soft)' }}>
+        <EmptyView>
          <Avatar
            src={currentAppInfo?.logo}
            size={80}
--- a/src/renderer/src/components/ObsidianExportDialog.tsx
+++ b/src/renderer/src/components/ObsidianExportDialog.tsx
@@ -38,7 +38,6 @@ interface PopupContainerProps {
  message?: Message
  messages?: Message[]
  topic?: Topic
-  rawContent?: string
 }

 // 转换文件信息数组为树形结构
@@ -141,8 +140,7 @@ const PopupContainer: React.FC<PopupContainerProps> = ({
  resolve,
  message,
  messages,
-  topic,
-  rawContent
+  topic
 }) => {
  const defaultObsidianVault = store.getState().settings.defaultObsidianVault
  const [state, setState] = useState({
@@ -231,9 +229,7 @@ const PopupContainer: React.FC<PopupContainerProps> = ({
      return
    }
    let markdown = ''
-    if (rawContent) {
-      markdown = rawContent
-    } else if (topic) {
+    if (topic) {
      markdown = await topicToMarkdown(topic, exportReasoning)
    } else if (messages && messages.length > 0) {
      markdown = messagesToMarkdown(messages, exportReasoning)
@@ -303,6 +299,7 @@ const PopupContainer: React.FC<PopupContainerProps> = ({
      }
    }
  }
+
  return (
    <Modal
      title={i18n.t('chat.topics.export.obsidian_atributes')}
@@ -413,11 +410,9 @@ const PopupContainer: React.FC<PopupContainerProps> = ({
            </Option>
          </Select>
        </Form.Item>
-        {!rawContent && (
-          <Form.Item label={i18n.t('chat.topics.export.obsidian_reasoning')}>
-            <Switch checked={exportReasoning} onChange={setExportReasoning} />
-          </Form.Item>
-        )}
+        <Form.Item label={i18n.t('chat.topics.export.obsidian_reasoning')}>
+          <Switch checked={exportReasoning} onChange={setExportReasoning} />
+        </Form.Item>
      </Form>
    </Modal>
  )
--- a/src/renderer/src/components/Popups/ObsidianExportPopup.tsx
+++ b/src/renderer/src/components/Popups/ObsidianExportPopup.tsx
@@ -9,7 +9,6 @@ interface ObsidianExportOptions {
  topic?: Topic
  message?: Message
  messages?: Message[]
-  rawContent?: string
 }

 export default class ObsidianExportPopup {
@@ -25,7 +24,6 @@ export default class ObsidianExportPopup {
          topic={options.topic}
          message={options.message}
          messages={options.messages}
-          rawContent={options.rawContent}
          obsidianTags={''}
          open={true}
          resolve={(v) => {
--- a/src/renderer/src/components/Popups/SaveToKnowledgePopup.tsx
+++ b/src/renderer/src/components/Popups/SaveToKnowledgePopup.tsx
@@ -253,39 +253,12 @@ const PopupContainer: React.FC<Props> = ({ source, title, resolve }) => {
    let savedCount = 0

    try {
-      // Validate knowledge base configuration before proceeding
-      if (!selectedBaseId) {
-        throw new Error('No knowledge base selected')
-      }
-
-      const selectedBase = bases.find((base) => base.id === selectedBaseId)
-      if (!selectedBase) {
-        throw new Error('Selected knowledge base not found')
-      }
-
-      if (!selectedBase.version) {
-        throw new Error('Knowledge base is not properly configured. Please check the knowledge base settings.')
-      }
-
      if (isNoteMode) {
        const note = source.data as NotesTreeNode
-        if (!note.externalPath) {
-          throw new Error('Note external path is required for export')
-        }
-
-        let content = ''
-        try {
-          content = await window.api.file.readExternal(note.externalPath)
-        } catch (error) {
-          logger.error('Failed to read note file:', error as Error)
-          throw new Error('Failed to read note content. Please ensure the file exists and is accessible.')
-        }
-
-        if (!content || content.trim() === '') {
-          throw new Error('Note content is empty. Cannot export empty notes to knowledge base.')
-        }
-
-        logger.debug('Note content loaded', { contentLength: content.length })
+        const content = note.externalPath
+          ? await window.api.file.readExternal(note.externalPath)
+          : await window.api.file.read(note.id + '.md')
+        logger.debug('Note content:', content)
        await addNote(content)
        savedCount = 1
      } else {
@@ -310,23 +283,9 @@ const PopupContainer: React.FC<Props> = ({ source, title, resolve }) => {
      resolve({ success: true, savedCount })
    } catch (error) {
      logger.error('save failed:', error as Error)
-
-      // Provide more specific error messages
-      let errorMessage = t(
-        isTopicMode ? 'chat.save.topic.knowledge.error.save_failed' : 'chat.save.knowledge.error.save_failed'
+      window.toast.error(
+        t(isTopicMode ? 'chat.save.topic.knowledge.error.save_failed' : 'chat.save.knowledge.error.save_failed')
      )
-
-      if (error instanceof Error) {
-        if (error.message.includes('not properly configured')) {
-          errorMessage = error.message
-        } else if (error.message.includes('empty')) {
-          errorMessage = error.message
-        } else if (error.message.includes('read note content')) {
-          errorMessage = error.message
-        }
-      }
-
-      window.toast.error(errorMessage)
      setLoading(false)
    }
  }
--- a/src/renderer/src/components/Popups/TextFilePreview.tsx
+++ b/src/renderer/src/components/Popups/TextFilePreview.tsx
@@ -55,15 +55,12 @@ const PopupContainer: React.FC<Props> = ({ text, title, extension, resolve }) =>
      footer={null}>
      {extension !== undefined ? (
        <Editor
-          readOnly={true}
+          editable={false}
          expanded={false}
          height="100%"
          style={{ height: '100%' }}
          value={text}
          language={extension}
-          options={{
-            keymap: true
-          }}
        />
      ) : (
        <Text>{text}</Text>
--- a/src/renderer/src/components/RichEditor/index.tsx
+++ b/src/renderer/src/components/RichEditor/index.tsx
@@ -48,8 +48,7 @@ const RichEditor = ({
  enableContentSearch = false,
  isFullWidth = false,
  fontFamily = 'default',
-  fontSize = 16,
-  enableSpellCheck = false
+  fontSize = 16
  // toolbarItems: _toolbarItems // TODO: Implement custom toolbar items
 }: RichEditorProps & { ref?: React.RefObject<RichEditorRef | null> }) => {
  // Use the rich editor hook for complete editor management
@@ -72,7 +71,6 @@ const RichEditor = ({
    onBlur,
    placeholder,
    editable,
-    enableSpellCheck,
    scrollParent: () => scrollContainerRef.current,
    onShowTableActionMenu: ({ position, actions }) => {
      const iconMap: Record<string, React.ReactNode> = {
--- a/src/renderer/src/components/RichEditor/styles.ts
+++ b/src/renderer/src/components/RichEditor/styles.ts
@@ -14,31 +14,6 @@ export const RichEditorWrapper = styled.div<{
  border-radius: 6px;
  background: var(--color-background);
  overflow-y: hidden;
-  .ProseMirror table,
-  .tiptap table {
-    table-layout: auto !important;
-  }
-
-  .ProseMirror table th,
-  .ProseMirror table td,
-  .tiptap th,
-  .tiptap td {
-    white-space: normal !important;
-    word-wrap: break-word !important;
-    word-break: break-word !important;
-    overflow-wrap: break-word !important;
-    overflow: visible !important;
-    text-overflow: clip !important;
-  }
-
-  .ProseMirror table th > *,
-  .ProseMirror table td > *,
-  .tiptap td > *,
-  .tiptap th > * {
-    white-space: normal !important;
-    overflow: visible !important;
-    text-overflow: clip !important;
-  }
  width: ${({ $isFullWidth }) => ($isFullWidth ? '100%' : '60%')};
  margin: ${({ $isFullWidth }) => ($isFullWidth ? '0' : '0 auto')};
  font-family: ${({ $fontFamily }) => ($fontFamily === 'serif' ? 'var(--font-family-serif)' : 'var(--font-family)')};
@@ -46,7 +21,6 @@ export const RichEditorWrapper = styled.div<{

  ${({ $minHeight }) => $minHeight && `min-height: ${$minHeight}px;`}
  ${({ $maxHeight }) => $maxHeight && `max-height: ${$maxHeight}px;`}
-
 `

 export const ToolbarWrapper = styled.div`
--- a/src/renderer/src/components/RichEditor/types.ts
+++ b/src/renderer/src/components/RichEditor/types.ts
@@ -50,8 +50,6 @@ export interface RichEditorProps {
  fontFamily?: 'default' | 'serif'
  /** Font size in pixels */
  fontSize?: number
-  /** Whether to enable spell check */
-  enableSpellCheck?: boolean
 }

 export interface ToolbarItem {
--- a/src/renderer/src/components/RichEditor/useRichEditor.ts
+++ b/src/renderer/src/components/RichEditor/useRichEditor.ts
@@ -57,8 +57,6 @@ export interface UseRichEditorOptions {
  editable?: boolean
  /** Whether to enable table of contents functionality */
  enableTableOfContents?: boolean
-  /** Whether to enable spell check */
-  enableSpellCheck?: boolean
  /** Show table action menu (row/column) with concrete actions and position */
  onShowTableActionMenu?: (payload: {
    type: 'row' | 'column'
@@ -128,7 +126,6 @@ export const useRichEditor = (options: UseRichEditorOptions = {}): UseRichEditor
    previewLength = 50,
    placeholder = '',
    editable = true,
-    enableSpellCheck = false,
    onShowTableActionMenu,
    scrollParent
  } = options
@@ -413,9 +410,7 @@ export const useRichEditor = (options: UseRichEditorOptions = {}): UseRichEditor
        // Allow text selection even when not editable
        style: editable
          ? ''
-          : 'user-select: text; -webkit-user-select: text; -moz-user-select: text; -ms-user-select: text;',
-        // Set spellcheck attribute on the contenteditable element
-        spellcheck: enableSpellCheck ? 'true' : 'false'
+          : 'user-select: text; -webkit-user-select: text; -moz-user-select: text; -ms-user-select: text;'
      }
    },
    onUpdate: ({ editor }) => {
--- a/src/renderer/src/components/Tab/TabContainer.tsx
+++ b/src/renderer/src/components/Tab/TabContainer.tsx
@@ -237,17 +237,7 @@ const TabsContainer: React.FC<TabsContainerProps> = ({ children }) => {
            onSortEnd={onSortEnd}
            className="tabs-sortable"
            renderItem={(tab) => (
-              <Tab
-                key={tab.id}
-                active={tab.id === activeTabId}
-                onClick={() => handleTabClick(tab)}
-                onAuxClick={(e) => {
-                  if (e.button === 1 && tab.id !== 'home') {
-                    e.preventDefault()
-                    e.stopPropagation()
-                    closeTab(tab.id)
-                  }
-                }}>
+              <Tab key={tab.id} active={tab.id === activeTabId} onClick={() => handleTabClick(tab)}>
                <TabHeader>
                  {tab.id && <TabIcon>{getTabIcon(tab.id, minapps, minAppsCache)}</TabIcon>}
                  <TabTitle>{getTabTitle(tab.id)}</TabTitle>
--- a/src/renderer/src/components/UpdateDialog.tsx
+++ b/src/renderer/src/components/UpdateDialog.tsx
@@ -1,101 +0,0 @@
-import { Button, Modal, ModalBody, ModalContent, ModalFooter, ModalHeader, ScrollShadow } from '@heroui/react'
-import { loggerService } from '@logger'
-import { handleSaveData } from '@renderer/store'
-import { ReleaseNoteInfo, UpdateInfo } from 'builder-util-runtime'
-import React, { useEffect, useState } from 'react'
-import { useTranslation } from 'react-i18next'
-import Markdown from 'react-markdown'
-
-const logger = loggerService.withContext('UpdateDialog')
-
-interface UpdateDialogProps {
-  isOpen: boolean
-  onClose: () => void
-  releaseInfo: UpdateInfo | null
-}
-
-const UpdateDialog: React.FC<UpdateDialogProps> = ({ isOpen, onClose, releaseInfo }) => {
-  const { t } = useTranslation()
-  const [isInstalling, setIsInstalling] = useState(false)
-
-  useEffect(() => {
-    if (isOpen && releaseInfo) {
-      logger.info('Update dialog opened', { version: releaseInfo.version })
-    }
-  }, [isOpen, releaseInfo])
-
-  const handleInstall = async () => {
-    setIsInstalling(true)
-    try {
-      await handleSaveData()
-      await window.api.quitAndInstall()
-    } catch (error) {
-      logger.error('Failed to save data before update', error as Error)
-      setIsInstalling(false)
-      window.toast.error(t('update.saveDataError'))
-    }
-  }
-
-  const releaseNotes = releaseInfo?.releaseNotes
-
-  return (
-    <Modal
-      isOpen={isOpen}
-      onClose={onClose}
-      size="2xl"
-      scrollBehavior="inside"
-      classNames={{
-        base: 'max-h-[85vh]',
-        header: 'border-b border-divider',
-        footer: 'border-t border-divider'
-      }}>
-      <ModalContent>
-        {(onModalClose) => (
-          <>
-            <ModalHeader className="flex flex-col gap-1">
-              <h3 className="font-semibold text-lg">{t('update.title')}</h3>
-              <p className="text-default-500 text-small">
-                {t('update.message').replace('{{version}}', releaseInfo?.version || '')}
-              </p>
-            </ModalHeader>
-
-            <ModalBody>
-              <ScrollShadow className="max-h-[450px]" hideScrollBar>
-                <div className="markdown rounded-lg bg-default-50 p-4">
-                  <Markdown>
-                    {typeof releaseNotes === 'string'
-                      ? releaseNotes
-                      : Array.isArray(releaseNotes)
-                        ? releaseNotes
-                            .map((note: ReleaseNoteInfo) => note.note)
-                            .filter(Boolean)
-                            .join('\n\n')
-                        : t('update.noReleaseNotes')}
-                  </Markdown>
-                </div>
-              </ScrollShadow>
-            </ModalBody>
-
-            <ModalFooter>
-              <Button variant="light" onPress={onModalClose} isDisabled={isInstalling}>
-                {t('update.later')}
-              </Button>
-
-              <Button
-                color="primary"
-                onPress={async () => {
-                  await handleInstall()
-                  onModalClose()
-                }}
-                isLoading={isInstalling}>
-                {t('update.install')}
-              </Button>
-            </ModalFooter>
-          </>
-        )}
-      </ModalContent>
-    </Modal>
-  )
-}
-
-export default UpdateDialog
--- a/src/renderer/src/config/minapps.ts
+++ b/src/renderer/src/config/minapps.ts
@@ -39,7 +39,6 @@ import PoeAppLogo from '@renderer/assets/images/apps/poe.webp?url'
 import QwenlmAppLogo from '@renderer/assets/images/apps/qwenlm.webp?url'
 import SensetimeAppLogo from '@renderer/assets/images/apps/sensetime.png?url'
 import SparkDeskAppLogo from '@renderer/assets/images/apps/sparkdesk.webp?url'
-import StepfunAppLogo from '@renderer/assets/images/apps/stepfun.png?url'
 import ThinkAnyLogo from '@renderer/assets/images/apps/thinkany.webp?url'
 import TiangongAiLogo from '@renderer/assets/images/apps/tiangong.png?url'
 import WanZhiAppLogo from '@renderer/assets/images/apps/wanzhi.jpg?url'
@@ -47,6 +46,7 @@ import WPSLingXiLogo from '@renderer/assets/images/apps/wpslingxi.webp?url'
 import XiaoYiAppLogo from '@renderer/assets/images/apps/xiaoyi.webp?url'
 import YouLogo from '@renderer/assets/images/apps/you.jpg?url'
 import TencentYuanbaoAppLogo from '@renderer/assets/images/apps/yuanbao.webp?url'
+import YuewenAppLogo from '@renderer/assets/images/apps/yuewen.png?url'
 import ZaiAppLogo from '@renderer/assets/images/apps/zai.png?url'
 import ZhihuAppLogo from '@renderer/assets/images/apps/zhihu.png?url'
 import ClaudeAppLogo from '@renderer/assets/images/models/claude.png?url'
@@ -145,14 +145,14 @@ const ORIGIN_DEFAULT_MIN_APPS: MinAppType[] = [
  {
    id: 'dashscope',
    name: i18n.t('minapps.qwen'),
-    url: 'https://www.tongyi.com/',
+    url: 'https://tongyi.aliyun.com/qianwen/',
    logo: QwenModelLogo
  },
  {
    id: 'stepfun',
-    name: i18n.t('minapps.stepfun'),
-    url: 'https://stepfun.com',
-    logo: StepfunAppLogo,
+    name: i18n.t('minapps.yuewen'),
+    url: 'https://yuewen.cn/chats/new',
+    logo: YuewenAppLogo,
    bodered: true
  },
  {
--- a/src/renderer/src/config/models/default.ts
+++ b/src/renderer/src/config/models/default.ts
@@ -25,7 +25,7 @@ export const SYSTEM_MODELS: Record<SystemProviderId | 'defaultModel', Model[]> =
    // Default quick assistant model
    glm45FlashModel
  ],
-  cherryin: [],
+  // cherryin: [],
  vertexai: [],
  '302ai': [
    {
@@ -260,7 +260,6 @@ export const SYSTEM_MODELS: Record<SystemProviderId | 'defaultModel', Model[]> =
    { id: 'deepseek-r1', name: 'DeepSeek-R1', provider: 'burncloud', group: 'deepseek-ai' },
    { id: 'deepseek-v3', name: 'DeepSeek-V3', provider: 'burncloud', group: 'deepseek-ai' }
  ],
-  ovms: [],
  ollama: [],
  lmstudio: [],
  silicon: [
@@ -430,12 +429,6 @@ export const SYSTEM_MODELS: Record<SystemProviderId | 'defaultModel', Model[]> =
    }
  ],
  anthropic: [
-    {
-      id: 'claude-sonnet-4-5-20250929',
-      provider: 'anthropic',
-      name: 'Claude Sonnet 4.5',
-      group: 'Claude 4.5'
-    },
    {
      id: 'claude-sonnet-4-20250514',
      provider: 'anthropic',
@@ -704,12 +697,6 @@ export const SYSTEM_MODELS: Record<SystemProviderId | 'defaultModel', Model[]> =
      name: 'GLM-4.5-Flash',
      group: 'GLM-4.5'
    },
-    {
-      id: 'glm-4.6',
-      provider: 'zhipu',
-      name: 'GLM-4.6',
-      group: 'GLM-4.6'
-    },
    {
      id: 'glm-4.5',
      provider: 'zhipu',
@@ -1817,19 +1804,5 @@ export const SYSTEM_MODELS: Record<SystemProviderId | 'defaultModel', Model[]> =
      provider: 'aionly',
      group: 'gemini'
    }
-  ],
-  longcat: [
-    {
-      id: 'LongCat-Flash-Chat',
-      name: 'LongCat Flash Chat',
-      provider: 'longcat',
-      group: 'LongCat'
-    },
-    {
-      id: 'LongCat-Flash-Thinking',
-      name: 'LongCat Flash Thinking',
-      provider: 'longcat',
-      group: 'LongCat'
-    }
  ]
 }
--- a/src/renderer/src/config/models/logo.ts
+++ b/src/renderer/src/config/models/logo.ts
@@ -61,7 +61,6 @@ import ChatGPTImageModelLogo from '@renderer/assets/images/models/gpt_image_1.pn
 import ChatGPTo1ModelLogo from '@renderer/assets/images/models/gpt_o1.png'
 import GPT5ModelLogo from '@renderer/assets/images/models/gpt-5.png'
 import GPT5ChatModelLogo from '@renderer/assets/images/models/gpt-5-chat.png'
-import GPT5CodexModelLogo from '@renderer/assets/images/models/gpt-5-codex.png'
 import GPT5MiniModelLogo from '@renderer/assets/images/models/gpt-5-mini.png'
 import GPT5NanoModelLogo from '@renderer/assets/images/models/gpt-5-nano.png'
 import GrokModelLogo from '@renderer/assets/images/models/grok.png'
@@ -163,7 +162,6 @@ export function getModelLogo(modelId: string) {
    return undefined
  }

-  // key is regex
  const logoMap = {
    pixtral: isLight ? PixtralModelLogo : PixtralModelLogoDark,
    jina: isLight ? JinaModelLogo : JinaModelLogoDark,
@@ -179,7 +177,6 @@ export function getModelLogo(modelId: string) {
    'gpt-5-mini': GPT5MiniModelLogo,
    'gpt-5-nano': GPT5NanoModelLogo,
    'gpt-5-chat': GPT5ChatModelLogo,
-    'gpt-5-codex': GPT5CodexModelLogo,
    'gpt-5': GPT5ModelLogo,
    gpts: isLight ? ChatGPT4ModelLogo : ChatGPT4ModelLogoDark,
    'gpt-oss(?:-[\\w-]+)': isLight ? ChatGptModelLogo : ChatGptModelLogoDark,
@@ -289,7 +286,7 @@ export function getModelLogo(modelId: string) {
    longcat: LongCatAppLogo,
    bytedance: BytedanceModelLogo,
    '(V_1|V_1_TURBO|V_2|V_2A|V_2_TURBO|DESCRIBE|UPSCALE)': IdeogramModelLogo
-  } as const
+  }

  for (const key in logoMap) {
    const regex = new RegExp(key, 'i')
--- a/src/renderer/src/config/models/reasoning.ts
+++ b/src/renderer/src/config/models/reasoning.ts
@@ -22,7 +22,6 @@ export const MODEL_SUPPORTED_REASONING_EFFORT: ReasoningEffortConfig = {
  default: ['low', 'medium', 'high'] as const,
  o: ['low', 'medium', 'high'] as const,
  gpt5: ['minimal', 'low', 'medium', 'high'] as const,
-  gpt5_codex: ['low', 'medium', 'high'] as const,
  grok: ['low', 'high'] as const,
  gemini: ['low', 'medium', 'high', 'auto'] as const,
  gemini_pro: ['low', 'medium', 'high', 'auto'] as const,
@@ -41,7 +40,6 @@ export const MODEL_SUPPORTED_OPTIONS: ThinkingOptionConfig = {
  default: ['off', ...MODEL_SUPPORTED_REASONING_EFFORT.default] as const,
  o: MODEL_SUPPORTED_REASONING_EFFORT.o,
  gpt5: [...MODEL_SUPPORTED_REASONING_EFFORT.gpt5] as const,
-  gpt5_codex: MODEL_SUPPORTED_REASONING_EFFORT.gpt5_codex,
  grok: MODEL_SUPPORTED_REASONING_EFFORT.grok,
  gemini: ['off', ...MODEL_SUPPORTED_REASONING_EFFORT.gemini] as const,
  gemini_pro: MODEL_SUPPORTED_REASONING_EFFORT.gemini_pro,
@@ -57,13 +55,8 @@ export const MODEL_SUPPORTED_OPTIONS: ThinkingOptionConfig = {

 export const getThinkModelType = (model: Model): ThinkingModelType => {
  let thinkingModelType: ThinkingModelType = 'default'
-  const modelId = getLowerBaseModelName(model.id)
  if (isGPT5SeriesModel(model)) {
-    if (modelId.includes('codex')) {
-      thinkingModelType = 'gpt5_codex'
-    } else {
-      thinkingModelType = 'gpt5'
-    }
+    thinkingModelType = 'gpt5'
  } else if (isSupportedReasoningEffortOpenAIModel(model)) {
    thinkingModelType = 'o'
  } else if (isSupportedThinkingTokenGeminiModel(model)) {
@@ -178,13 +171,9 @@ export function isGeminiReasoningModel(model?: Model): boolean {
  return false
 }

-// Gemini 支持思考模式的模型正则
-export const GEMINI_THINKING_MODEL_REGEX =
-  /gemini-(?:2\.5.*(?:-latest)?|flash-latest|pro-latest|flash-lite-latest)(?:-[\w-]+)*$/i
-
 export const isSupportedThinkingTokenGeminiModel = (model: Model): boolean => {
  const modelId = getLowerBaseModelName(model.id, '/')
-  if (GEMINI_THINKING_MODEL_REGEX.test(modelId)) {
+  if (modelId.includes('gemini-2.5')) {
    if (modelId.includes('image') || modelId.includes('tts')) {
      return false
    }
@@ -339,20 +328,14 @@ export const isSupportedReasoningEffortPerplexityModel = (model: Model): boolean

 export const isSupportedThinkingTokenZhipuModel = (model: Model): boolean => {
  const modelId = getLowerBaseModelName(model.id, '/')
-  return ['glm-4.5', 'glm-4.6'].some((id) => modelId.includes(id))
+  return modelId.includes('glm-4.5')
 }

 export const isDeepSeekHybridInferenceModel = (model: Model) => {
  const modelId = getLowerBaseModelName(model.id)
  // deepseek官方使用chat和reasoner做推理控制，其他provider需要单独判断，id可能会有所差别
  // openrouter: deepseek/deepseek-chat-v3.1 不知道会不会有其他provider仿照ds官方分出一个同id的作为非思考模式的模型，这里有风险
-  // Matches: "deepseek-v3" followed by ".digit" or "-digit".
-  // Optionally, this can be followed by ".alphanumeric_sequence" or "-alphanumeric_sequence"
-  // until the end of the string.
-  // Examples: deepseek-v3.1, deepseek-v3-1, deepseek-v3.1.2, deepseek-v3.1-alpha
-  // Does NOT match: deepseek-v3.123 (missing separator after '1'), deepseek-v3.x (x isn't a digit)
-  // TODO: move to utils and add test cases
-  return /deepseek-v3(?:\.\d|-\d)(?:(\.|-)\w+)?$/.test(modelId) || modelId.includes('deepseek-chat-v3.1')
+  return /deepseek-v3(?:\.1|-1-\d+)?/.test(modelId) || modelId.includes('deepseek-chat-v3.1')
 }

 export const isSupportedThinkingTokenDeepSeekModel = isDeepSeekHybridInferenceModel
--- a/src/renderer/src/config/models/vision.ts
+++ b/src/renderer/src/config/models/vision.ts
@@ -12,7 +12,6 @@ const visionAllowedModels = [
  'gemini-1\\.5',
  'gemini-2\\.0',
  'gemini-2\\.5',
-  'gemini-(flash|pro|flash-lite)-latest',
  'gemini-exp',
  'claude-3',
  'claude-sonnet-4',
@@ -22,9 +21,7 @@ const visionAllowedModels = [
  'qwen-vl',
  'qwen2-vl',
  'qwen2.5-vl',
-  'qwen3-vl',
  'qwen2.5-omni',
-  'qwen3-omni',
  'qvq',
  'internvl2',
  'grok-vision-beta',
--- a/src/renderer/src/config/models/websearch.ts
+++ b/src/renderer/src/config/models/websearch.ts
@@ -11,12 +11,9 @@ export const CLAUDE_SUPPORTED_WEBSEARCH_REGEX = new RegExp(
  'i'
 )

-export const GEMINI_FLASH_MODEL_REGEX = new RegExp('gemini.*-flash.*$')
+export const GEMINI_FLASH_MODEL_REGEX = new RegExp('gemini-.*-flash.*$')

-export const GEMINI_SEARCH_REGEX = new RegExp(
-  'gemini-(?:2.*(?:-latest)?|flash-latest|pro-latest|flash-lite-latest)(?:-[\\w-]+)*$',
-  'i'
-)
+export const GEMINI_SEARCH_REGEX = new RegExp('gemini-2\\..*', 'i')

 export const PERPLEXITY_SEARCH_MODELS = [
  'sonar-pro',
--- a/src/renderer/src/config/providers.ts
+++ b/src/renderer/src/config/providers.ts
@@ -24,11 +24,9 @@ import GrokProviderLogo from '@renderer/assets/images/providers/grok.png'
 import GroqProviderLogo from '@renderer/assets/images/providers/groq.png'
 import HyperbolicProviderLogo from '@renderer/assets/images/providers/hyperbolic.png'
 import InfiniProviderLogo from '@renderer/assets/images/providers/infini.png'
-import IntelOvmsLogo from '@renderer/assets/images/providers/intel.png'
 import JinaProviderLogo from '@renderer/assets/images/providers/jina.png'
 import LanyunProviderLogo from '@renderer/assets/images/providers/lanyun.png'
 import LMStudioProviderLogo from '@renderer/assets/images/providers/lmstudio.png'
-import LongCatProviderLogo from '@renderer/assets/images/providers/longcat.png'
 import MinimaxProviderLogo from '@renderer/assets/images/providers/minimax.png'
 import MistralProviderLogo from '@renderer/assets/images/providers/mistral.png'
 import ModelScopeProviderLogo from '@renderer/assets/images/providers/modelscope.png'
@@ -80,16 +78,16 @@ export const CHERRYAI_PROVIDER: SystemProvider = {
 }

 export const SYSTEM_PROVIDERS_CONFIG: Record<SystemProviderId, SystemProvider> = {
-  cherryin: {
-    id: 'cherryin',
-    name: 'CherryIN',
-    type: 'openai',
-    apiKey: '',
-    apiHost: 'https://open.cherryin.net',
-    models: [],
-    isSystem: true,
-    enabled: true
-  },
+  // cherryin: {
+  //   id: 'cherryin',
+  //   name: 'CherryIN',
+  //   type: 'openai',
+  //   apiKey: '',
+  //   apiHost: 'https://open.cherryin.ai',
+  //   models: [],
+  //   isSystem: true,
+  //   enabled: true
+  // },
  silicon: {
    id: 'silicon',
    name: 'Silicon',
@@ -110,16 +108,6 @@ export const SYSTEM_PROVIDERS_CONFIG: Record<SystemProviderId, SystemProvider> =
    isSystem: true,
    enabled: false
  },
-  ovms: {
-    id: 'ovms',
-    name: 'OpenVINO Model Server',
-    type: 'openai',
-    apiKey: '',
-    apiHost: 'http://localhost:8000/v3/',
-    models: SYSTEM_MODELS.ovms,
-    isSystem: true,
-    enabled: false
-  },
  ocoolai: {
    id: 'ocoolai',
    name: 'ocoolAI',
@@ -634,16 +622,6 @@ export const SYSTEM_PROVIDERS_CONFIG: Record<SystemProviderId, SystemProvider> =
    models: SYSTEM_MODELS['poe'],
    isSystem: true,
    enabled: false
-  },
-  longcat: {
-    id: 'longcat',
-    name: 'LongCat',
-    type: 'openai',
-    apiKey: '',
-    apiHost: 'https://api.longcat.chat/openai',
-    models: SYSTEM_MODELS.longcat,
-    isSystem: true,
-    enabled: false
  }
 } as const

@@ -660,7 +638,6 @@ export const PROVIDER_LOGO_MAP: AtLeast<SystemProviderId, string> = {
  yi: ZeroOneProviderLogo,
  groq: GroqProviderLogo,
  zhipu: ZhipuProviderLogo,
-  ovms: IntelOvmsLogo,
  ollama: OllamaProviderLogo,
  lmstudio: LMStudioProviderLogo,
  moonshot: MoonshotProviderLogo,
@@ -707,8 +684,7 @@ export const PROVIDER_LOGO_MAP: AtLeast<SystemProviderId, string> = {
  'new-api': NewAPIProviderLogo,
  'aws-bedrock': AwsProviderLogo,
  poe: 'poe', // use svg icon component
-  aionly: AiOnlyProviderLogo,
-  longcat: LongCatProviderLogo
+  aionly: AiOnlyProviderLogo
 } as const

 export function getProviderLogo(providerId: string) {
@@ -732,17 +708,17 @@ type ProviderUrls = {
 }

 export const PROVIDER_URLS: Record<SystemProviderId, ProviderUrls> = {
-  cherryin: {
-    api: {
-      url: 'https://open.cherryin.net'
-    },
-    websites: {
-      official: 'https://open.cherryin.ai',
-      apiKey: 'https://open.cherryin.ai/console/token',
-      docs: 'https://open.cherryin.ai',
-      models: 'https://open.cherryin.ai/pricing'
-    }
-  },
+  // cherryin: {
+  //   api: {
+  //     url: 'https://open.cherryin.ai'
+  //   },
+  //   websites: {
+  //     official: 'https://open.cherryin.ai',
+  //     apiKey: 'https://open.cherryin.ai/console/token',
+  //     docs: 'https://open.cherryin.ai',
+  //     models: 'https://open.cherryin.ai/pricing'
+  //   }
+  // },
  ph8: {
    api: {
      url: 'https://ph8.co'
@@ -1046,16 +1022,6 @@ export const PROVIDER_URLS: Record<SystemProviderId, ProviderUrls> = {
      models: 'https://console.groq.com/docs/models'
    }
  },
-  ovms: {
-    api: {
-      url: 'http://localhost:8000/v3/'
-    },
-    websites: {
-      official: 'https://www.intel.com/content/www/us/en/developer/tools/openvino-toolkit/overview.html',
-      docs: 'https://docs.openvino.ai/2025/model-server/ovms_what_is_openvino_model_server.html',
-      models: 'https://www.modelscope.cn/organization/OpenVINO'
-    }
-  },
  ollama: {
    api: {
      url: 'http://localhost:11434'
@@ -1324,17 +1290,6 @@ export const PROVIDER_URLS: Record<SystemProviderId, ProviderUrls> = {
      docs: 'https://www.aiionly.com/document',
      models: 'https://www.aiionly.com'
    }
-  },
-  longcat: {
-    api: {
-      url: 'https://api.longcat.chat/openai'
-    },
-    websites: {
-      official: 'https://longcat.chat',
-      apiKey: 'https://longcat.chat/platform/api_keys',
-      docs: 'https://longcat.chat/platform/docs/zh/',
-      models: 'https://longcat.chat/platform/docs/zh/APIDocs.html'
-    }
  }
 }

--- a/src/renderer/src/databases/index.ts
+++ b/src/renderer/src/databases/index.ts
@@ -7,6 +7,7 @@ import {
 } from '@renderer/types'
 // Import necessary types for blocks and new message structure
 import type { Message as NewMessage, MessageBlock } from '@renderer/types/newMessage'
+import { NotesTreeNode } from '@renderer/types/note'
 import { Dexie, type EntityTable } from 'dexie'

 import { upgradeToV5, upgradeToV7, upgradeToV8 } from './upgrades'
@@ -23,6 +24,7 @@ export const db = new Dexie('CherryStudio', {
  quick_phrases: EntityTable<QuickPhrase, 'id'>
  message_blocks: EntityTable<MessageBlock, 'id'> // Correct type for message_blocks
  translate_languages: EntityTable<CustomTranslateLanguage, 'id'>
+  notes_tree: EntityTable<{ id: string; tree: NotesTreeNode[] }, 'id'>
 }

 db.version(1).stores({
@@ -116,7 +118,8 @@ db.version(10).stores({
  translate_history: '&id, sourceText, targetText, sourceLanguage, targetLanguage, createdAt',
  translate_languages: '&id, langCode',
  quick_phrases: 'id',
-  message_blocks: 'id, messageId, file.id'
+  message_blocks: 'id, messageId, file.id',
+  notes_tree: '&id'
 })

 export default db
--- a/src/renderer/src/hooks/useCodeTools.ts
+++ b/src/renderer/src/hooks/useCodeTools.ts
@@ -108,11 +108,7 @@ export const useCodeTools = () => {
  const environmentVariables = codeToolsState?.environmentVariables?.[codeToolsState.selectedCliTool] || ''

  // 检查是否可以启动（所有必需字段都已填写）
-  const canLaunch = Boolean(
-    codeToolsState.selectedCliTool &&
-      codeToolsState.currentDirectory &&
-      (codeToolsState.selectedCliTool === codeTools.githubCopilotCli || selectedModel)
-  )
+  const canLaunch = Boolean(codeToolsState.selectedCliTool && selectedModel && codeToolsState.currentDirectory)

  return {
    // 状态
--- a/src/renderer/src/hooks/useKnowledge.ts
+++ b/src/renderer/src/hooks/useKnowledge.ts
@@ -360,7 +360,7 @@ export const useKnowledgeBases = () => {
  const deleteKnowledgeBase = (baseId: string) => {
    const base = bases.find((b) => b.id === baseId)
    if (!base) return
-    dispatch(deleteBase({ baseId }))
+    dispatch(deleteBase({ baseId, baseParams: getKnowledgeBaseParams(base) }))

    // remove assistant knowledge_base
    const _assistants = assistants.map((assistant) => {
--- a/src/renderer/src/hooks/useTopic.ts
+++ b/src/renderer/src/hooks/useTopic.ts
@@ -48,17 +48,6 @@ export function useActiveTopic(assistantId: string, topic?: Topic) {
    }
  }, [activeTopic?.id, assistant])

-  useEffect(() => {
-    if (!assistant?.topics?.length || !activeTopic) {
-      return
-    }
-
-    const latestTopic = assistant.topics.find((item) => item.id === activeTopic.id)
-    if (latestTopic && latestTopic !== activeTopic) {
-      setActiveTopic(latestTopic)
-    }
-  }, [assistant?.topics, activeTopic])
-
  return { activeTopic, setActiveTopic }
 }

--- a/src/renderer/src/i18n/label.ts
+++ b/src/renderer/src/i18n/label.ts
@@ -61,7 +61,6 @@ const providerKeyMap = {
  nvidia: 'provider.nvidia',
  o3: 'provider.o3',
  ocoolai: 'provider.ocoolai',
-  ovms: 'provider.ovms',
  ollama: 'provider.ollama',
  openai: 'provider.openai',
  openrouter: 'provider.openrouter',
--- a/src/renderer/src/i18n/locales/en-us.json
+++ b/src/renderer/src/i18n/locales/en-us.json
@@ -251,7 +251,6 @@
    "added": "Added",
    "case_sensitive": "Case Sensitive",
    "collapse": "Collapse",
-    "download": "Download",
    "includes_user_questions": "Include Your Questions",
    "manage": "Manage",
    "select_model": "Select Model",
@@ -1583,13 +1582,13 @@
    "nami-ai-search": "Nami AI Search",
    "qwen": "Qwen",
    "sensechat": "SenseChat",
-    "stepfun": "Stepfun",
    "tencent-yuanbao": "Yuanbao",
    "tiangong-ai": "Skywork",
    "wanzhi": "Wanzhi",
    "wenxin": "ERNIE",
    "wps-copilot": "WPS Copilot",
    "xiaoyi": "Xiaoyi",
+    "yuewen": "Yuewen",
    "zhihu": "Zhihu"
  },
  "miniwindow": {
@@ -1697,12 +1696,6 @@
    "provider_settings": "Go to provider settings"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "Note is empty, cannot generate name",
-      "failed": "Failed to generate note name",
-      "label": "Generate Note Name",
-      "success": "Note name generated successfully"
-    },
    "characters": "Characters",
    "collapse": "Collapse",
    "content_placeholder": "Please enter the note content...",
@@ -1784,8 +1777,6 @@
    "sort_updated_asc": "Update time (oldest first)",
    "sort_updated_desc": "Update time (newest first)",
    "sort_z2a": "File name (Z-A)",
-    "spell_check": "Spell Check",
-    "spell_check_tooltip": "Enable/Disable spell check",
    "star": "Favorite note",
    "starred_notes": "Collected notes",
    "title": "Notes",
@@ -1835,57 +1826,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "Install",
-      "installing": "Installing",
-      "reinstall": "Re-Install",
-      "run": "Run OVMS",
-      "starting": "Starting",
-      "stop": "Stop OVMS",
-      "stopping": "Stopping"
-    },
-    "description": "<div><p>1. Download OV Models.</p><p>2. Add Models in 'Manager'.</p><p>Support Windows Only!</p><p>OVMS Install Path: '%USERPROFILE%\\.cherrystudio\\ovms' .</p><p>Please refer to <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Intel OVMS Guide</a></p></dev>",
-    "download": {
-      "button": "Download",
-      "error": "Download Error",
-      "model_id": {
-        "label": "Model ID:",
-        "model_id_pattern": "Model ID must start with OpenVINO/",
-        "placeholder": "Required e.g. OpenVINO/Qwen3-8B-int4-ov",
-        "required": "Please enter the model ID"
-      },
-      "model_name": {
-        "label": "Model Name:",
-        "placeholder": "Required e.g. Qwen3-8B-int4-ov",
-        "required": "Please enter the model name"
-      },
-      "model_source": "Model Source:",
-      "model_task": "Model Task:",
-      "success": "Download successful",
-      "success_desc": "Model \"{{modelName}}\"-\"{{modelId}}\" downloaded successfully, please go to the OVMS management interface to add the model",
-      "tip": "The model is downloading, sometimes it takes hours. Please be patient...",
-      "title": "Download Intel OpenVINO Model"
-    },
-    "failed": {
-      "install": "Install OVMS failed:",
-      "install_code_100": "Unknown Error",
-      "install_code_101": "Only supports Intel(R) Core(TM) Ultra CPU",
-      "install_code_102": "Only supports Windows",
-      "install_code_103": "Download OVMS runtime failed",
-      "install_code_104": "Uncompress OVMS runtime failed",
-      "install_code_105": "Clean OVMS runtime failed",
-      "run": "Run OVMS failed:",
-      "stop": "Stop OVMS failed:"
-    },
-    "status": {
-      "not_installed": "OVMS is not installed",
-      "not_running": "OVMS is not running",
-      "running": "OVMS is running",
-      "unknown": "OVMS status unknown"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "Aspect Ratio",
    "aspect_ratios": {
@@ -2117,7 +2057,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexity",
    "ph8": "PH8",
    "poe": "Poe",
@@ -4412,7 +4351,6 @@
    "later": "Later",
    "message": "New version {{version}} is ready, do you want to install it now?",
    "noReleaseNotes": "No release notes",
-    "saveDataError": "Failed to save data, please try again.",
    "title": "Update"
  },
  "warning": {
--- a/src/renderer/src/i18n/locales/zh-cn.json
+++ b/src/renderer/src/i18n/locales/zh-cn.json
@@ -251,7 +251,6 @@
    "added": "已添加",
    "case_sensitive": "区分大小写",
    "collapse": "收起",
-    "download": "下载",
    "includes_user_questions": "包含用户提问",
    "manage": "管理",
    "select_model": "选择模型",
@@ -1583,13 +1582,13 @@
    "nami-ai-search": "纳米AI搜索",
    "qwen": "通义千问",
    "sensechat": "商量",
-    "stepfun": "阶跃AI",
    "tencent-yuanbao": "腾讯元宝",
    "tiangong-ai": "天工AI",
    "wanzhi": "万知",
    "wenxin": "文心一言",
    "wps-copilot": "WPS灵犀",
    "xiaoyi": "小艺",
+    "yuewen": "跃问",
    "zhihu": "知乎直答"
  },
  "miniwindow": {
@@ -1697,12 +1696,6 @@
    "provider_settings": "跳转到服务商设置界面"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "笔记为空，无法生成名称",
-      "failed": "生成笔记名称失败",
-      "label": "生成笔记名称",
-      "success": "笔记名称生成成功"
-    },
    "characters": "字符",
    "collapse": "收起",
    "content_placeholder": "请输入笔记内容...",
@@ -1784,8 +1777,6 @@
    "sort_updated_asc": "更新时间（从旧到新）",
    "sort_updated_desc": "更新时间（从新到旧）",
    "sort_z2a": "文件名（Z-A）",
-    "spell_check": "拼写检查",
-    "spell_check_tooltip": "启用/禁用拼写检查",
    "star": "收藏笔记",
    "starred_notes": "收藏的笔记",
    "title": "笔记",
@@ -1835,57 +1826,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "安装",
-      "installing": "正在安装",
-      "reinstall": "重装",
-      "run": "运行 OVMS",
-      "starting": "启动中",
-      "stop": "停止 OVMS",
-      "stopping": "停止中"
-    },
-    "description": "<div><p>1. 下载 OV 模型.</p><p>2. 在 'Manager' 中添加模型.</p><p>仅支持 Windows!</p><p>OVMS 安装路径: '%USERPROFILE%\\.cherrystudio\\ovms' .</p><p>请参考 <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Intel OVMS 指南</a></p></dev>",
-    "download": {
-      "button": "下载",
-      "error": "选择失败",
-      "model_id": {
-        "label": "模型 ID",
-        "model_id_pattern": "模型 ID 必须以 OpenVINO/ 开头",
-        "placeholder": "必填，例如 OpenVINO/Qwen3-8B-int4-ov",
-        "required": "请输入模型 ID"
-      },
-      "model_name": {
-        "label": "模型名称",
-        "placeholder": "必填，例如 Qwen3-8B-int4-ov",
-        "required": "请输入模型名称"
-      },
-      "model_source": "模型来源:",
-      "model_task": "模型任务:",
-      "success": "下载成功",
-      "success_desc": "模型\"{{modelName}}\"-\"{{modelId}}\"下载成功，请前往 OVMS 管理界面添加模型",
-      "tip": "模型正在下载，有时需要几个小时。请耐心等待...",
-      "title": "下载 Intel OpenVINO 模型"
-    },
-    "failed": {
-      "install": "安装 OVMS 失败:",
-      "install_code_100": "未知错误",
-      "install_code_101": "仅支持 Intel(R) Core(TM) Ultra CPU",
-      "install_code_102": "仅支持 Windows",
-      "install_code_103": "下载 OVMS runtime 失败",
-      "install_code_104": "解压 OVMS runtime 失败",
-      "install_code_105": "清理 OVMS runtime 失败",
-      "run": "运行 OVMS 失败:",
-      "stop": "停止 OVMS 失败:"
-    },
-    "status": {
-      "not_installed": "OVMS 未安装",
-      "not_running": "OVMS 未运行",
-      "running": "OVMS 正在运行",
-      "unknown": "OVMS 状态未知"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "画幅比例",
    "aspect_ratios": {
@@ -2117,7 +2057,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexity",
    "ph8": "PH8 大模型开放平台",
    "poe": "Poe",
@@ -4412,7 +4351,6 @@
    "later": "稍后",
    "message": "发现新版本 {{version}}，是否立即安装？",
    "noReleaseNotes": "暂无更新日志",
-    "saveDataError": "保存数据失败，请重试",
    "title": "更新提示"
  },
  "warning": {
--- a/src/renderer/src/i18n/locales/zh-tw.json
+++ b/src/renderer/src/i18n/locales/zh-tw.json
@@ -251,7 +251,6 @@
    "added": "已新增",
    "case_sensitive": "區分大小寫",
    "collapse": "折疊",
-    "download": "下載",
    "includes_user_questions": "包含使用者提問",
    "manage": "管理",
    "select_model": "選擇模型",
@@ -1583,13 +1582,13 @@
    "nami-ai-search": "納米AI搜索",
    "qwen": "通義千問",
    "sensechat": "商量",
-    "stepfun": "階躍AI",
    "tencent-yuanbao": "騰訊元寶",
    "tiangong-ai": "天工AI",
    "wanzhi": "萬知",
    "wenxin": "文心一言",
    "wps-copilot": "WPS靈犀",
    "xiaoyi": "小藝",
+    "yuewen": "躍問",
    "zhihu": "知乎直答"
  },
  "miniwindow": {
@@ -1697,12 +1696,6 @@
    "provider_settings": "跳轉到服務商設置界面"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "筆記為空，無法生成名稱",
-      "failed": "生成筆記名稱失敗",
-      "label": "生成筆記名稱",
-      "success": "筆記名稱生成成功"
-    },
    "characters": "字符",
    "collapse": "收起",
    "content_placeholder": "請輸入筆記內容...",
@@ -1784,8 +1777,6 @@
    "sort_updated_asc": "更新時間（從舊到新）",
    "sort_updated_desc": "更新時間（從新到舊）",
    "sort_z2a": "文件名（Z-A）",
-    "spell_check": "拼寫檢查",
-    "spell_check_tooltip": "啟用/禁用拼寫檢查",
    "star": "收藏筆記",
    "starred_notes": "收藏的筆記",
    "title": "筆記",
@@ -1835,57 +1826,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "安裝",
-      "installing": "正在安裝",
-      "reinstall": "重新安裝",
-      "run": "執行 OVMS",
-      "starting": "啟動中",
-      "stop": "停止 OVMS",
-      "stopping": "停止中"
-    },
-    "description": "<div><p>1. 下載 OV 模型。</p><p>2. 在 'Manager' 中新增模型。</p><p>僅支援 Windows！</p><p>OVMS 安裝路徑: '%USERPROFILE%\\.cherrystudio\\ovms' 。</p><p>請參考 <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Intel OVMS 指南</a></p></dev>",
-    "download": {
-      "button": "下載",
-      "error": "選擇失敗",
-      "model_id": {
-        "label": "模型 ID",
-        "model_id_pattern": "模型 ID 必須以 OpenVINO/ 開頭",
-        "placeholder": "必填，例如 OpenVINO/Qwen3-8B-int4-ov",
-        "required": "請輸入模型 ID"
-      },
-      "model_name": {
-        "label": "模型名稱",
-        "placeholder": "必填，例如 Qwen3-8B-int4-ov",
-        "required": "請輸入模型名稱"
-      },
-      "model_source": "模型來源:",
-      "model_task": "模型任務:",
-      "success": "下載成功",
-      "success_desc": "模型\"{{modelName}}\"-\"{{modelId}}\"下載成功，請前往 OVMS 管理界面添加模型",
-      "tip": "模型正在下載，有時需要幾個小時。請耐心等候...",
-      "title": "下載 Intel OpenVINO 模型"
-    },
-    "failed": {
-      "install": "安裝 OVMS 失敗:",
-      "install_code_100": "未知錯誤",
-      "install_code_101": "僅支援 Intel(R) Core(TM) Ultra CPU",
-      "install_code_102": "僅支援 Windows",
-      "install_code_103": "下載 OVMS runtime 失敗",
-      "install_code_104": "解壓 OVMS runtime 失敗",
-      "install_code_105": "清理 OVMS runtime 失敗",
-      "run": "執行 OVMS 失敗:",
-      "stop": "停止 OVMS 失敗:"
-    },
-    "status": {
-      "not_installed": "OVMS 未安裝",
-      "not_running": "OVMS 未執行",
-      "running": "OVMS 正在執行",
-      "unknown": "OVMS 狀態未知"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "畫幅比例",
    "aspect_ratios": {
@@ -2117,7 +2057,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexity",
    "ph8": "PH8 大模型開放平台",
    "poe": "Poe",
@@ -4412,7 +4351,6 @@
    "later": "稍後",
    "message": "新版本 {{version}} 已準備就緒，是否立即安裝？",
    "noReleaseNotes": "暫無更新日誌",
-    "saveDataError": "保存數據失敗，請重試",
    "title": "更新提示"
  },
  "warning": {
--- a/src/renderer/src/i18n/translate/el-gr.json
+++ b/src/renderer/src/i18n/translate/el-gr.json
@@ -251,7 +251,6 @@
    "added": "προστέθηκε",
    "case_sensitive": "Διάκριση πεζών/κεφαλαίων",
    "collapse": "συμπεριλάβετε",
-    "download": "Λήψη",
    "includes_user_questions": "Περιλαμβάνει ερωτήσεις χρήστη",
    "manage": "χειριστείτε",
    "select_model": "επιλογή μοντέλου",
@@ -334,7 +333,6 @@
      "new_topic": "Νέο θέμα {{Command}}",
      "pause": "Παύση",
      "placeholder": "Εισάγετε μήνυμα εδώ...",
-      "placeholder_without_triggers": "Εδώ εισαγάγετε το μήνυμα, πατήστε {{key}} για αποστολή",
      "send": "Αποστολή",
      "settings": "Ρυθμίσεις",
      "thinking": {
@@ -1583,13 +1581,13 @@
    "nami-ai-search": "Nami AI Search",
    "qwen": "Qwen",
    "sensechat": "SenseChat",
-    "stepfun": "Stepfun",
    "tencent-yuanbao": "Yuanbao",
    "tiangong-ai": "Skywork",
    "wanzhi": "Wanzhi",
    "wenxin": "ERNIE",
    "wps-copilot": "WPS Copilot",
    "xiaoyi": "Xiaoyi",
+    "yuewen": "Yuewen",
    "zhihu": "Zhihu"
  },
  "miniwindow": {
@@ -1697,12 +1695,6 @@
    "provider_settings": "Μετάβαση στις ρυθμίσεις παρόχου"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "Το σημείωμα είναι κενό, δεν μπορεί να δημιουργηθεί όνομα",
-      "failed": "Αποτυχία δημιουργίας ονόματος σημείωσης",
-      "label": "Δημιουργία ονόματος σημείωσης",
-      "success": "Η δημιουργία του ονόματος σημειώσεων ολοκληρώθηκε με επιτυχία"
-    },
    "characters": "χαρακτήρας",
    "collapse": "σύμπτυξη",
    "content_placeholder": "Παρακαλώ εισαγάγετε το περιεχόμενο των σημειώσεων...",
@@ -1784,8 +1776,6 @@
    "sort_updated_asc": "χρόνος ενημέρωσης (από παλιά στα νέα)",
    "sort_updated_desc": "χρόνος ενημέρωσης (από νεώτερο σε παλαιότερο)",
    "sort_z2a": "όνομα αρχείου (Z-A)",
-    "spell_check": "Έλεγχος ορθογραφίας",
-    "spell_check_tooltip": "Ενεργοποίηση/Απενεργοποίηση ελέγχου ορθογραφίας",
    "star": "Αγαπημένες σημειώσεις",
    "starred_notes": "Σημειώσεις συλλογής",
    "title": "σημειώσεις",
@@ -1835,57 +1825,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "Εγκατάσταση",
-      "installing": "Εγκατάσταση σε εξέλιξη",
-      "reinstall": "Επανεγκατάσταση",
-      "run": "Εκτέλεση OVMS",
-      "starting": "Εκκίνηση σε εξέλιξη",
-      "stop": "Διακοπή OVMS",
-      "stopping": "Διακοπή σε εξέλιξη"
-    },
-    "description": "<div><p>1. Λήψη μοντέλου OV.</p><p>2. Προσθήκη μοντέλου στο 'Manager'.</p><p>Υποστηρίζεται μόνο στα Windows!</p><p>Διαδρομή εγκατάστασης OVMS: '%USERPROFILE%\\.cherrystudio\\ovms' .</p><p>Ανατρέξτε στον <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Οδηγό Intel OVMS</a></p></div>",
-    "download": {
-      "button": "Λήψη",
-      "error": "Η επιλογή απέτυχε",
-      "model_id": {
-        "label": "Αναγνωριστικό μοντέλου:",
-        "model_id_pattern": "Το αναγνωριστικό μοντέλου πρέπει να ξεκινά με OpenVINO/",
-        "placeholder": "Απαιτείται, π.χ. OpenVINO/Qwen3-8B-int4-ov",
-        "required": "Παρακαλώ εισάγετε το αναγνωριστικό μοντέλου"
-      },
-      "model_name": {
-        "label": "Όνομα μοντέλου:",
-        "placeholder": "Απαιτείται, π.χ. Qwen3-8B-int4-ov",
-        "required": "Παρακαλώ εισάγετε το όνομα του μοντέλου"
-      },
-      "model_source": "Πηγή μοντέλου:",
-      "model_task": "Εργασία μοντέλου:",
-      "success": "Η λήψη ολοκληρώθηκε με επιτυχία",
-      "success_desc": "Το μοντέλο \"{{modelName}}\"-\"{{modelId}}\" λήφθηκε επιτυχώς, παρακαλώ μεταβείτε στη διεπαφή διαχείρισης OVMS για να προσθέσετε το μοντέλο",
-      "tip": "Το μοντέλο κατεβαίνει, μερικές φορές χρειάζονται αρκετές ώρες. Παρακαλώ περιμένετε υπομονετικά...",
-      "title": "Λήψη μοντέλου Intel OpenVINO"
-    },
-    "failed": {
-      "install": "Η εγκατάσταση του OVMS απέτυχε:",
-      "install_code_100": "Άγνωστο σφάλμα",
-      "install_code_101": "Υποστηρίζεται μόνο σε Intel(R) Core(TM) Ultra CPU",
-      "install_code_102": "Υποστηρίζεται μόνο στα Windows",
-      "install_code_103": "Η λήψη του OVMS runtime απέτυχε",
-      "install_code_104": "Η αποσυμπίεση του OVMS runtime απέτυχε",
-      "install_code_105": "Ο καθαρισμός του OVMS runtime απέτυχε",
-      "run": "Η εκτέλεση του OVMS απέτυχε:",
-      "stop": "Η διακοπή του OVMS απέτυχε:"
-    },
-    "status": {
-      "not_installed": "Το OVMS δεν έχει εγκατασταθεί",
-      "not_running": "Το OVMS δεν εκτελείται",
-      "running": "Το OVMS εκτελείται",
-      "unknown": "Άγνωστη κατάσταση OVMS"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "Λόγος διαστάσεων",
    "aspect_ratios": {
@@ -2117,7 +2056,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexity",
    "ph8": "Πλατφόρμα Ανοιχτής Μεγάλης Μοντέλου PH8",
    "poe": "Poe",
@@ -4412,7 +4350,6 @@
    "later": "Μετά",
    "message": "Νέα έκδοση {{version}} είναι έτοιμη, θέλετε να την εγκαταστήσετε τώρα;",
    "noReleaseNotes": "Χωρίς σημειώσεις",
-    "saveDataError": "Η αποθήκευση των δεδομένων απέτυχε, δοκιμάστε ξανά",
    "title": "Ενημέρωση"
  },
  "warning": {
--- a/src/renderer/src/i18n/translate/es-es.json
+++ b/src/renderer/src/i18n/translate/es-es.json
@@ -251,7 +251,6 @@
    "added": "Agregado",
    "case_sensitive": "Distingue mayúsculas y minúsculas",
    "collapse": "Colapsar",
-    "download": "Descargar",
    "includes_user_questions": "Incluye preguntas del usuario",
    "manage": "Administrar",
    "select_model": "Seleccionar Modelo",
@@ -334,7 +333,6 @@
      "new_topic": "Nuevo tema {{Command}}",
      "pause": "Pausar",
      "placeholder": "Escribe aquí tu mensaje...",
-      "placeholder_without_triggers": "Escriba un mensaje aquí y presione {{key}} para enviar",
      "send": "Enviar",
      "settings": "Configuración",
      "thinking": {
@@ -1583,13 +1581,13 @@
    "nami-ai-search": "Nami AI Search",
    "qwen": "Qwen",
    "sensechat": "SenseChat",
-    "stepfun": "Stepfun",
    "tencent-yuanbao": "Yuanbao",
    "tiangong-ai": "Skywork",
    "wanzhi": "Wanzhi",
    "wenxin": "ERNIE",
    "wps-copilot": "WPS Copilot",
    "xiaoyi": "Xiaoyi",
+    "yuewen": "Yuewen",
    "zhihu": "Zhihu"
  },
  "miniwindow": {
@@ -1697,12 +1695,6 @@
    "provider_settings": "Ir a la configuración del proveedor"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "La nota está vacía, no se puede generar un nombre",
-      "failed": "Error al generar el nombre de la nota",
-      "label": "Generar nombre de nota",
-      "success": "Se ha generado correctamente el nombre de la nota"
-    },
    "characters": "carácter",
    "collapse": "ocultar",
    "content_placeholder": "Introduzca el contenido de la nota...",
@@ -1784,8 +1776,6 @@
    "sort_updated_asc": "Fecha de actualización (de más antigua a más reciente)",
    "sort_updated_desc": "Fecha de actualización (de más nuevo a más antiguo)",
    "sort_z2a": "Nombre de archivo (Z-A)",
-    "spell_check": "comprobación ortográfica",
-    "spell_check_tooltip": "Habilitar/deshabilitar revisión ortográfica",
    "star": "Notas guardadas",
    "starred_notes": "notas guardadas",
    "title": "notas",
@@ -1835,57 +1825,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "Instalar",
-      "installing": "Instalando",
-      "reinstall": "Reinstalar",
-      "run": "Ejecutar OVMS",
-      "starting": "Iniciando",
-      "stop": "Detener OVMS",
-      "stopping": "Deteniendo"
-    },
-    "description": "<div><p>1. Descargar modelo OV.</p><p>2. Agregar modelo en 'Administrador'.</p><p>¡Solo compatible con Windows!</p><p>Ruta de instalación de OVMS: '%USERPROFILE%\\.cherrystudio\\ovms' .</p><p>Consulte la <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Guía de Intel OVMS</a></p></dev>",
-    "download": {
-      "button": "Descargar",
-      "error": "Selección fallida",
-      "model_id": {
-        "label": "ID del modelo:",
-        "model_id_pattern": "El ID del modelo debe comenzar con OpenVINO/",
-        "placeholder": "Requerido, por ejemplo, OpenVINO/Qwen3-8B-int4-ov",
-        "required": "Por favor, ingrese el ID del modelo"
-      },
-      "model_name": {
-        "label": "Nombre del modelo:",
-        "placeholder": "Requerido, por ejemplo, Qwen3-8B-int4-ov",
-        "required": "Por favor, ingrese el nombre del modelo"
-      },
-      "model_source": "Fuente del modelo:",
-      "model_task": "Tarea del modelo:",
-      "success": "Descarga exitosa",
-      "success_desc": "El modelo \"{{modelName}}\"-\"{{modelId}}\" se descargó exitosamente, por favor vaya a la interfaz de administración de OVMS para agregar el modelo",
-      "tip": "El modelo se está descargando, a veces toma varias horas. Por favor espere pacientemente...",
-      "title": "Descargar modelo Intel OpenVINO"
-    },
-    "failed": {
-      "install": "Error al instalar OVMS:",
-      "install_code_100": "Error desconocido",
-      "install_code_101": "Solo compatible con CPU Intel(R) Core(TM) Ultra",
-      "install_code_102": "Solo compatible con Windows",
-      "install_code_103": "Error al descargar el tiempo de ejecución de OVMS",
-      "install_code_104": "Error al descomprimir el tiempo de ejecución de OVMS",
-      "install_code_105": "Error al limpiar el tiempo de ejecución de OVMS",
-      "run": "Error al ejecutar OVMS:",
-      "stop": "Error al detener OVMS:"
-    },
-    "status": {
-      "not_installed": "OVMS no instalado",
-      "not_running": "OVMS no está en ejecución",
-      "running": "OVMS en ejecución",
-      "unknown": "Estado de OVMS desconocido"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "Relación de aspecto",
    "aspect_ratios": {
@@ -2117,7 +2056,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplejidad",
    "ph8": "Plataforma Abierta de Grandes Modelos PH8",
    "poe": "Poe",
@@ -4412,7 +4350,6 @@
    "later": "Más tarde",
    "message": "Nueva versión {{version}} disponible, ¿desea instalarla ahora?",
    "noReleaseNotes": "Sin notas de la versión",
-    "saveDataError": "Error al guardar los datos, inténtalo de nuevo",
    "title": "Actualización"
  },
  "warning": {
--- a/src/renderer/src/i18n/translate/fr-fr.json
+++ b/src/renderer/src/i18n/translate/fr-fr.json
@@ -251,7 +251,6 @@
    "added": "Ajouté",
    "case_sensitive": "Respecter la casse",
    "collapse": "Réduire",
-    "download": "Télécharger",
    "includes_user_questions": "Inclure les questions de l'utilisateur",
    "manage": "Gérer",
    "select_model": "Sélectionner le Modèle",
@@ -334,7 +333,6 @@
      "new_topic": "Nouveau sujet {{Command}}",
      "pause": "Pause",
      "placeholder": "Entrez votre message ici...",
-      "placeholder_without_triggers": "Entrez votre message ici, appuyez sur {{key}} pour envoyer",
      "send": "Envoyer",
      "settings": "Paramètres",
      "thinking": {
@@ -1583,13 +1581,13 @@
    "nami-ai-search": "Nami AI Search",
    "qwen": "Qwen",
    "sensechat": "SenseChat",
-    "stepfun": "Stepfun",
    "tencent-yuanbao": "Yuanbao",
    "tiangong-ai": "Skywork",
    "wanzhi": "Wanzhi",
    "wenxin": "ERNIE",
    "wps-copilot": "WPS Copilot",
    "xiaoyi": "Xiaoyi",
+    "yuewen": "Yuewen",
    "zhihu": "Zhihu"
  },
  "miniwindow": {
@@ -1697,12 +1695,6 @@
    "provider_settings": "Aller aux paramètres du fournisseur"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "La note est vide, impossible de générer un nom",
-      "failed": "Échec de la génération du nom de note",
-      "label": "Générer un nom de note",
-      "success": "La génération du nom de note a réussi"
-    },
    "characters": "caractère",
    "collapse": "réduire",
    "content_placeholder": "Veuillez saisir le contenu de la note...",
@@ -1784,8 +1776,6 @@
    "sort_updated_asc": "Heure de mise à jour (du plus ancien au plus récent)",
    "sort_updated_desc": "Date de mise à jour (du plus récent au plus ancien)",
    "sort_z2a": "Nom de fichier (Z-A)",
-    "spell_check": "Vérification orthographique",
-    "spell_check_tooltip": "Activer/Désactiver la vérification orthographique",
    "star": "Notes enregistrées",
    "starred_notes": "notes de collection",
    "title": "notes",
@@ -1835,57 +1825,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "Installer",
-      "installing": "Installation en cours",
-      "reinstall": "Réinstaller",
-      "run": "Exécuter OVMS",
-      "starting": "Démarrage en cours",
-      "stop": "Arrêter OVMS",
-      "stopping": "Arrêt en cours"
-    },
-    "description": "<div><p>1. Télécharger le modèle OV.</p><p>2. Ajouter le modèle dans 'Manager'.</p><p>Uniquement compatible avec Windows !</p><p>Chemin d'installation d'OVMS : '%USERPROFILE%\\.cherrystudio\\ovms' .</p><p>Veuillez vous référer au <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Guide Intel OVMS</a></p></dev>",
-    "download": {
-      "button": "Télécharger",
-      "error": "Échec de la sélection",
-      "model_id": {
-        "label": "ID du modèle :",
-        "model_id_pattern": "L'ID du modèle doit commencer par OpenVINO/",
-        "placeholder": "Requis, par exemple OpenVINO/Qwen3-8B-int4-ov",
-        "required": "Veuillez saisir l'ID du modèle"
-      },
-      "model_name": {
-        "label": "Nom du modèle :",
-        "placeholder": "Requis, par exemple Qwen3-8B-int4-ov",
-        "required": "Veuillez saisir le nom du modèle"
-      },
-      "model_source": "Source du modèle :",
-      "model_task": "Tâche du modèle :",
-      "success": "Téléchargement réussi",
-      "success_desc": "Le modèle \"{{modelName}}\"-\"{{modelId}}\" a été téléchargé avec succès, veuillez vous rendre à l'interface de gestion OVMS pour ajouter le modèle",
-      "tip": "Le modèle est en cours de téléchargement, cela peut parfois prendre plusieurs heures. Veuillez patienter...",
-      "title": "Télécharger le modèle Intel OpenVINO"
-    },
-    "failed": {
-      "install": "Échec de l'installation d'OVMS :",
-      "install_code_100": "Erreur inconnue",
-      "install_code_101": "Uniquement compatible avec les processeurs Intel(R) Core(TM) Ultra",
-      "install_code_102": "Uniquement compatible avec Windows",
-      "install_code_103": "Échec du téléchargement du runtime OVMS",
-      "install_code_104": "Échec de la décompression du runtime OVMS",
-      "install_code_105": "Échec du nettoyage du runtime OVMS",
-      "run": "Échec de l'exécution d'OVMS :",
-      "stop": "Échec de l'arrêt d'OVMS :"
-    },
-    "status": {
-      "not_installed": "OVMS non installé",
-      "not_running": "OVMS n'est pas en cours d'exécution",
-      "running": "OVMS en cours d'exécution",
-      "unknown": "État d'OVMS inconnu"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "Format d'image",
    "aspect_ratios": {
@@ -2117,7 +2056,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexité",
    "ph8": "Plateforme ouverte de grands modèles PH8",
    "poe": "Poe",
@@ -4412,7 +4350,6 @@
    "later": "Plus tard",
    "message": "Nouvelle version {{version}} disponible, voulez-vous l'installer maintenant ?",
    "noReleaseNotes": "Aucune note de version",
-    "saveDataError": "Échec de la sauvegarde des données, veuillez réessayer",
    "title": "Mise à jour"
  },
  "warning": {
--- a/src/renderer/src/i18n/translate/ja-jp.json
+++ b/src/renderer/src/i18n/translate/ja-jp.json
@@ -251,7 +251,6 @@
    "added": "追加済み",
    "case_sensitive": "大文字と小文字の区別",
    "collapse": "折りたたむ",
-    "download": "ダウンロード",
    "includes_user_questions": "ユーザーからの質問を含む",
    "manage": "管理",
    "select_model": "モデルを選択",
@@ -334,7 +333,6 @@
      "new_topic": "新しいトピック {{Command}}",
      "pause": "一時停止",
      "placeholder": "ここにメッセージを入力し、{{key}} を押して送信...",
-      "placeholder_without_triggers": "ここにメッセージを入力し、{{key}} を押して送信してください",
      "send": "送信",
      "settings": "設定",
      "thinking": {
@@ -1583,13 +1581,13 @@
    "nami-ai-search": "Nami AI Search",
    "qwen": "通義千問",
    "sensechat": "SenseChat",
-    "stepfun": "Stepfun",
    "tencent-yuanbao": "騰訊元宝",
    "tiangong-ai": "Skywork",
    "wanzhi": "万知",
    "wenxin": "ERNIE",
    "wps-copilot": "WPS Copilot",
    "xiaoyi": "小藝",
+    "yuewen": "躍問",
    "zhihu": "知乎直答"
  },
  "miniwindow": {
@@ -1697,12 +1695,6 @@
    "provider_settings": "プロバイダー設定に移動"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "ノートが空です。名前を生成できません。",
-      "failed": "ノート名の生成に失敗しました",
-      "label": "ノート名の生成",
-      "success": "ノート名の生成に成功しました"
-    },
    "characters": "文字",
    "collapse": "閉じる",
    "content_placeholder": "メモの内容を入力してください...",
@@ -1784,8 +1776,6 @@
    "sort_updated_asc": "更新日時（古い順）",
    "sort_updated_desc": "更新日時（新しい順）",
    "sort_z2a": "ファイル名（Z-A）",
-    "spell_check": "スペルチェック",
-    "spell_check_tooltip": "スペルチェックの有効/無効",
    "star": "お気に入りのノート",
    "starred_notes": "収集したノート",
    "title": "ノート",
@@ -1835,57 +1825,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "インストール",
-      "installing": "インストール中",
-      "reinstall": "再インストール",
-      "run": "OVMSを実行",
-      "starting": "起動中",
-      "stop": "OVMSを停止",
-      "stopping": "停止中"
-    },
-    "description": "<div><p>1. OVモデルをダウンロードします。</p><p>2. 'マネージャー'でモデルを追加します。</p><p>Windowsのみサポート！</p><p>OVMSインストールパス: '%USERPROFILE%\\.cherrystudio\\ovms' 。</p><p>詳細は<a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Intel OVMSガイド</a>をご参照ください。</p></dev>",
-    "download": {
-      "button": "ダウンロード",
-      "error": "ダウンロードエラー",
-      "model_id": {
-        "label": "モデルID",
-        "model_id_pattern": "モデルIDはOpenVINO/で始まる必要があります",
-        "placeholder": "必須 例: OpenVINO/Qwen3-8B-int4-ov",
-        "required": "モデルIDを入力してください"
-      },
-      "model_name": {
-        "label": "モデル名",
-        "placeholder": "必須 例: Qwen3-8B-int4-ov",
-        "required": "モデル名を入力してください"
-      },
-      "model_source": "モデルソース:",
-      "model_task": "モデルタスク:",
-      "success": "ダウンロード成功",
-      "success_desc": "モデル\"{{modelName}}\"-\"{{modelId}}\"ダウンロード成功、OVMS管理インターフェースに移動してモデルを追加してください",
-      "tip": "モデルはダウンロードされていますが、時には数時間かかります。我慢してください...",
-      "title": "Intel OpenVINOモデルをダウンロード"
-    },
-    "failed": {
-      "install": "OVMSのインストールに失敗しました:",
-      "install_code_100": "不明なエラー",
-      "install_code_101": "Intel(R) Core(TM) Ultra CPUのみサポート",
-      "install_code_102": "Windowsのみサポート",
-      "install_code_103": "OVMSランタイムのダウンロードに失敗しました",
-      "install_code_104": "OVMSランタイムの解凍に失敗しました",
-      "install_code_105": "OVMSランタイムのクリーンアップに失敗しました",
-      "run": "OVMSの実行に失敗しました:",
-      "stop": "OVMSの停止に失敗しました:"
-    },
-    "status": {
-      "not_installed": "OVMSはインストールされていません",
-      "not_running": "OVMSは実行されていません",
-      "running": "OVMSは実行中です",
-      "unknown": "OVMSのステータスが不明です"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "画幅比例",
    "aspect_ratios": {
@@ -2117,7 +2056,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexity",
    "ph8": "PH8",
    "poe": "Poe",
@@ -4412,7 +4350,6 @@
    "later": "後で",
    "message": "新バージョン {{version}} が利用可能です。今すぐインストールしますか？",
    "noReleaseNotes": "暫無更新日誌",
-    "saveDataError": "データの保存に失敗しました。もう一度お試しください。",
    "title": "更新"
  },
  "warning": {
--- a/src/renderer/src/i18n/translate/pt-pt.json
+++ b/src/renderer/src/i18n/translate/pt-pt.json
@@ -251,7 +251,6 @@
    "added": "Adicionado",
    "case_sensitive": "Diferenciar maiúsculas e minúsculas",
    "collapse": "Recolher",
-    "download": "Baixar",
    "includes_user_questions": "Incluir perguntas do usuário",
    "manage": "Gerenciar",
    "select_model": "Selecionar Modelo",
@@ -334,7 +333,6 @@
      "new_topic": "Novo tópico {{Command}}",
      "pause": "Pausar",
      "placeholder": "Digite sua mensagem aqui...",
-      "placeholder_without_triggers": "Digite a mensagem aqui, pressione {{key}} para enviar",
      "send": "Enviar",
      "settings": "Configurações",
      "thinking": {
@@ -1583,13 +1581,13 @@
    "nami-ai-search": "Nami AI Search",
    "qwen": "Qwen",
    "sensechat": "SenseChat",
-    "stepfun": "Stepfun",
    "tencent-yuanbao": "Yuanbao",
    "tiangong-ai": "Skywork",
    "wanzhi": "Wanzhi",
    "wenxin": "ERNIE",
    "wps-copilot": "WPS Copilot",
    "xiaoyi": "Xiaoyi",
+    "yuewen": "Yuewen",
    "zhihu": "Zhihu"
  },
  "miniwindow": {
@@ -1697,12 +1695,6 @@
    "provider_settings": "Ir para as configurações do provedor"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "A nota está vazia, não é possível gerar um nome",
-      "failed": "Falha ao gerar o nome da nota",
-      "label": "Gerar nome da nota",
-      "success": "Nome da nota gerado com sucesso"
-    },
    "characters": "caractere",
    "collapse": "[minimizar]",
    "content_placeholder": "Introduza o conteúdo da nota...",
@@ -1784,8 +1776,6 @@
    "sort_updated_asc": "Tempo de atualização (do mais antigo para o mais recente)",
    "sort_updated_desc": "atualização de tempo (do mais novo para o mais antigo)",
    "sort_z2a": "Nome do arquivo (Z-A)",
-    "spell_check": "verificação ortográfica",
-    "spell_check_tooltip": "Ativar/Desativar verificação ortográfica",
    "star": "Notas favoritas",
    "starred_notes": "notas salvas",
    "title": "nota",
@@ -1835,57 +1825,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "Instalar",
-      "installing": "Instalando",
-      "reinstall": "Reinstalar",
-      "run": "Executar OVMS",
-      "starting": "Iniciando",
-      "stop": "Parar OVMS",
-      "stopping": "Parando"
-    },
-    "description": "<div><p>1. Baixe o modelo OV.</p><p>2. Adicione o modelo no 'Gerenciador'.</p><p>Compatível apenas com Windows!</p><p>Caminho de instalação do OVMS: '%USERPROFILE%\\.cherrystudio\\ovms' .</p><p>Consulte o <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>Guia do Intel OVMS</a></p></dev>",
-    "download": {
-      "button": "Baixar",
-      "error": "Falha na seleção",
-      "model_id": {
-        "label": "ID do modelo:",
-        "model_id_pattern": "O ID do modelo deve começar com OpenVINO/",
-        "placeholder": "Obrigatório, por exemplo, OpenVINO/Qwen3-8B-int4-ov",
-        "required": "Por favor, insira o ID do modelo"
-      },
-      "model_name": {
-        "label": "Nome do modelo:",
-        "placeholder": "Obrigatório, por exemplo, Qwen3-8B-int4-ov",
-        "required": "Por favor, insira o nome do modelo"
-      },
-      "model_source": "Fonte do modelo:",
-      "model_task": "Tarefa do modelo:",
-      "success": "Download concluído com sucesso",
-      "success_desc": "O modelo \"{{modelName}}\"-\"{{modelId}}\" foi baixado com sucesso, por favor vá para a interface de gerenciamento OVMS para adicionar o modelo",
-      "tip": "O modelo está sendo baixado, às vezes leva várias horas. Por favor aguarde pacientemente...",
-      "title": "Baixar modelo Intel OpenVINO"
-    },
-    "failed": {
-      "install": "Falha na instalação do OVMS:",
-      "install_code_100": "Erro desconhecido",
-      "install_code_101": "Compatível apenas com CPU Intel(R) Core(TM) Ultra",
-      "install_code_102": "Compatível apenas com Windows",
-      "install_code_103": "Falha ao baixar o tempo de execução do OVMS",
-      "install_code_104": "Falha ao descompactar o tempo de execução do OVMS",
-      "install_code_105": "Falha ao limpar o tempo de execução do OVMS",
-      "run": "Falha ao executar o OVMS:",
-      "stop": "Falha ao parar o OVMS:"
-    },
-    "status": {
-      "not_installed": "OVMS não instalado",
-      "not_running": "OVMS não está em execução",
-      "running": "OVMS em execução",
-      "unknown": "Status do OVMS desconhecido"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "Proporção da Imagem",
    "aspect_ratios": {
@@ -2117,7 +2056,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexidade",
    "ph8": "Plataforma Aberta de Grandes Modelos PH8",
    "poe": "Poe",
@@ -4412,7 +4350,6 @@
    "later": "Mais tarde",
    "message": "Nova versão {{version}} disponível, deseja instalar agora?",
    "noReleaseNotes": "Sem notas de versão",
-    "saveDataError": "Falha ao salvar os dados, tente novamente",
    "title": "Atualização"
  },
  "warning": {
--- a/src/renderer/src/i18n/translate/ru-ru.json
+++ b/src/renderer/src/i18n/translate/ru-ru.json
@@ -251,7 +251,6 @@
    "added": "Добавлено",
    "case_sensitive": "Чувствительность к регистру",
    "collapse": "Свернуть",
-    "download": "Скачать",
    "includes_user_questions": "Включает вопросы пользователей",
    "manage": "Редактировать",
    "select_model": "Выбрать модель",
@@ -334,7 +333,6 @@
      "new_topic": "Новый топик {{Command}}",
      "pause": "Остановить",
      "placeholder": "Введите ваше сообщение здесь, нажмите {{key}} для отправки...",
-      "placeholder_without_triggers": "Введите сообщение здесь, нажмите {{key}}, чтобы отправить",
      "send": "Отправить",
      "settings": "Настройки",
      "thinking": {
@@ -1583,13 +1581,13 @@
    "nami-ai-search": "Nami AI Search",
    "qwen": "Qwen",
    "sensechat": "SenseChat",
-    "stepfun": "Stepfun",
    "tencent-yuanbao": "Tencent Yuanbao",
    "tiangong-ai": "Skywork",
    "wanzhi": "Wanzhi",
    "wenxin": "ERNIE",
    "wps-copilot": "WPS Copilot",
    "xiaoyi": "Xiaoyi",
+    "yuewen": "Yuewen",
    "zhihu": "Zhihu"
  },
  "miniwindow": {
@@ -1697,12 +1695,6 @@
    "provider_settings": "Перейти к настройкам поставщика"
  },
  "notes": {
-    "auto_rename": {
-      "empty_note": "Заметки пусты, имя невозможно сгенерировать",
-      "failed": "Создание названия заметки не удалось",
-      "label": "Создать название заметки",
-      "success": "Имя заметки успешно создано"
-    },
    "characters": "Символы",
    "collapse": "Свернуть",
    "content_placeholder": "Введите содержимое заметки...",
@@ -1784,8 +1776,6 @@
    "sort_updated_asc": "Время обновления (от старого к новому)",
    "sort_updated_desc": "Время обновления (от нового к старому)",
    "sort_z2a": "Имя файла (Я-А)",
-    "spell_check": "Проверка орфографии",
-    "spell_check_tooltip": "Включить/отключить проверку орфографии",
    "star": "Избранные заметки",
    "starred_notes": "Сохраненные заметки",
    "title": "заметки",
@@ -1835,57 +1825,6 @@
    },
    "title": "Ollama"
  },
-  "ovms": {
-    "action": {
-      "install": "Установить",
-      "installing": "Установка",
-      "reinstall": "Переустановить",
-      "run": "Запустить OVMS",
-      "starting": "Запуск",
-      "stop": "Остановить OVMS",
-      "stopping": "Остановка"
-    },
-    "description": "<div><p>1. Загрузите модели OV.</p><p>2. Добавьте модели в 'Менеджер'.</p><p>Поддерживается только Windows!</p><p>Путь установки OVMS: '%USERPROFILE%\\.cherrystudio\\ovms'.</p><p>Пожалуйста, ознакомьтесь с <a href=https://github.com/openvinotoolkit/model_server/blob/c55551763d02825829337b62c2dcef9339706f79/docs/deploying_server_baremetal.md>руководством Intel OVMS</a></p></dev>",
-    "download": {
-      "button": "Скачать",
-      "error": "Ошибка загрузки",
-      "model_id": {
-        "label": "ID модели",
-        "model_id_pattern": "ID модели должен начинаться с OpenVINO/",
-        "placeholder": "Обязательно, например: OpenVINO/Qwen3-8B-int4-ov",
-        "required": "Пожалуйста, введите ID модели"
-      },
-      "model_name": {
-        "label": "Название модели:",
-        "placeholder": "Обязательно, например: Qwen3-8B-int4-ov",
-        "required": "Пожалуйста, введите название модели"
-      },
-      "model_source": "Источник модели:",
-      "model_task": "Задача модели:",
-      "success": "Скачивание успешно",
-      "success_desc": "Модель \"{{modelName}}\"-\"{{modelId}}\" успешно скачана, пожалуйста, перейдите в интерфейс управления OVMS, чтобы добавить модель",
-      "tip": "Модель загружается, иногда это занимает часы. Пожалуйста, будьте терпеливы...",
-      "title": "Скачать модель Intel OpenVINO"
-    },
-    "failed": {
-      "install": "Ошибка установки OVMS:",
-      "install_code_100": "Неизвестная ошибка",
-      "install_code_101": "Поддерживаются только процессоры Intel(R) Core(TM) Ultra CPU",
-      "install_code_102": "Поддерживается только Windows",
-      "install_code_103": "Ошибка загрузки среды выполнения OVMS",
-      "install_code_104": "Ошибка распаковки среды выполнения OVMS",
-      "install_code_105": "Ошибка очистки среды выполнения OVMS",
-      "run": "Ошибка запуска OVMS:",
-      "stop": "Ошибка остановки OVMS:"
-    },
-    "status": {
-      "not_installed": "OVMS не установлен",
-      "not_running": "OVMS не запущен",
-      "running": "OVMS запущен",
-      "unknown": "Статус OVMS неизвестен"
-    },
-    "title": "Intel OVMS"
-  },
  "paintings": {
    "aspect_ratio": "Пропорции изображения",
    "aspect_ratios": {
@@ -2117,7 +2056,6 @@
    "ollama": "Ollama",
    "openai": "OpenAI",
    "openrouter": "OpenRouter",
-    "ovms": "Intel OVMS",
    "perplexity": "Perplexity",
    "ph8": "PH8",
    "poe": "Poe",
@@ -4412,7 +4350,6 @@
    "later": "Позже",
    "message": "Новая версия {{version}} готова, установить сейчас?",
    "noReleaseNotes": "Нет заметок об обновлении",
-    "saveDataError": "Ошибка сохранения данных, повторите попытку",
    "title": "Обновление"
  },
  "warning": {
--- a/src/renderer/src/pages/code/CodeToolsPage.tsx
+++ b/src/renderer/src/pages/code/CodeToolsPage.tsx
@@ -98,10 +98,6 @@ const CodeToolsPage: FC = () => {
        return m.id.includes('openai') || OPENAI_CODEX_SUPPORTED_PROVIDERS.includes(m.provider)
      }

-      if (selectedCliTool === codeTools.githubCopilotCli) {
-        return false
-      }
-
      if (selectedCliTool === codeTools.qwenCode || selectedCliTool === codeTools.iFlowCli) {
        if (m.supported_endpoint_types) {
          return ['openai', 'openai-response'].some((type) =>
@@ -200,7 +196,7 @@ const CodeToolsPage: FC = () => {
      }
    }

-    if (!selectedModel && selectedCliTool !== codeTools.githubCopilotCli) {
+    if (!selectedModel) {
      return { isValid: false, message: t('code.model_required') }
    }

@@ -209,11 +205,6 @@ const CodeToolsPage: FC = () => {

  // 准备启动环境
  const prepareLaunchEnvironment = async (): Promise<Record<string, string> | null> => {
-    if (selectedCliTool === codeTools.githubCopilotCli) {
-      const userEnv = parseEnvironmentVariables(environmentVariables)
-      return userEnv
-    }
-
    if (!selectedModel) return null

    const modelProvider = getProviderByModel(selectedModel)
@@ -238,9 +229,7 @@ const CodeToolsPage: FC = () => {

  // 执行启动操作
  const executeLaunch = async (env: Record<string, string>) => {
-    const modelId = selectedCliTool === codeTools.githubCopilotCli ? '' : selectedModel?.id!
-
-    window.api.codeTools.run(selectedCliTool, modelId, currentDirectory, env, {
+    window.api.codeTools.run(selectedCliTool, selectedModel?.id!, currentDirectory, env, {
      autoUpdateToLatest,
      terminal: selectedTerminal
    })
@@ -327,12 +316,7 @@ const CodeToolsPage: FC = () => {
                banner
                style={{ borderRadius: 'var(--list-item-border-radius)' }}
                message={
-                  <div
-                    style={{
-                      display: 'flex',
-                      justifyContent: 'space-between',
-                      alignItems: 'center'
-                    }}>
+                  <div style={{ display: 'flex', justifyContent: 'space-between', alignItems: 'center' }}>
                    <span>{t('code.bun_required_message')}</span>
                    <Button
                      type="primary"
@@ -361,64 +345,46 @@ const CodeToolsPage: FC = () => {
              />
            </SettingsItem>

-            {selectedCliTool !== codeTools.githubCopilotCli && (
-              <SettingsItem>
-                <div className="settings-label">
-                  {t('code.model')}
-                  {selectedCliTool === 'claude-code' && (
-                    <Popover
-                      content={
-                        <div style={{ width: 200 }}>
-                          <div style={{ marginBottom: 8, fontWeight: 500 }}>{t('code.supported_providers')}</div>
-                          <div
-                            style={{
-                              display: 'flex',
-                              flexDirection: 'column',
-                              gap: 8
-                            }}>
-                            {getClaudeSupportedProviders(allProviders).map((provider) => {
-                              return (
-                                <Link
-                                  key={provider.id}
-                                  style={{
-                                    color: 'var(--color-text)',
-                                    display: 'flex',
-                                    alignItems: 'center',
-                                    gap: 4
-                                  }}
-                                  to={`/settings/provider?id=${provider.id}`}>
-                                  <ProviderLogo shape="square" src={getProviderLogo(provider.id)} size={20} />
-                                  {getProviderLabel(provider.id)}
-                                  <ArrowUpRight size={14} />
-                                </Link>
-                              )
-                            })}
-                          </div>
+            <SettingsItem>
+              <div className="settings-label">
+                {t('code.model')}
+                {selectedCliTool === 'claude-code' && (
+                  <Popover
+                    content={
+                      <div style={{ width: 200 }}>
+                        <div style={{ marginBottom: 8, fontWeight: 500 }}>{t('code.supported_providers')}</div>
+                        <div style={{ display: 'flex', flexDirection: 'column', gap: 8 }}>
+                          {getClaudeSupportedProviders(allProviders).map((provider) => {
+                            return (
+                              <Link
+                                key={provider.id}
+                                style={{ color: 'var(--color-text)', display: 'flex', alignItems: 'center', gap: 4 }}
+                                to={`/settings/provider?id=${provider.id}`}>
+                                <ProviderLogo shape="square" src={getProviderLogo(provider.id)} size={20} />
+                                {getProviderLabel(provider.id)}
+                                <ArrowUpRight size={14} />
+                              </Link>
+                            )
+                          })}
                        </div>
-                      }
-                      trigger="hover"
-                      placement="right">
-                      <HelpCircle
-                        size={14}
-                        style={{
-                          color: 'var(--color-text-3)',
-                          cursor: 'pointer'
-                        }}
-                      />
-                    </Popover>
-                  )}
-                </div>
-                <ModelSelector
-                  providers={availableProviders}
-                  predicate={modelPredicate}
-                  style={{ width: '100%' }}
-                  placeholder={t('code.model_placeholder')}
-                  value={selectedModel ? getModelUniqId(selectedModel) : undefined}
-                  onChange={handleModelChange}
-                  allowClear
-                />
-              </SettingsItem>
-            )}
+                      </div>
+                    }
+                    trigger="hover"
+                    placement="right">
+                    <HelpCircle size={14} style={{ color: 'var(--color-text-3)', cursor: 'pointer' }} />
+                  </Popover>
+                )}
+              </div>
+              <ModelSelector
+                providers={availableProviders}
+                predicate={modelPredicate}
+                style={{ width: '100%' }}
+                placeholder={t('code.model_placeholder')}
+                value={selectedModel ? getModelUniqId(selectedModel) : undefined}
+                onChange={handleModelChange}
+                allowClear
+              />
+            </SettingsItem>

            <SettingsItem>
              <div className="settings-label">{t('code.working_directory')}</div>
@@ -437,27 +403,11 @@ const CodeToolsPage: FC = () => {
                  options={directories.map((dir) => ({
                    value: dir,
                    label: (
-                      <div
-                        style={{
-                          display: 'flex',
-                          justifyContent: 'space-between',
-                          alignItems: 'center'
-                        }}>
-                        <span
-                          style={{
-                            flex: 1,
-                            overflow: 'hidden',
-                            textOverflow: 'ellipsis'
-                          }}>
-                          {dir}
-                        </span>
+                      <div style={{ display: 'flex', justifyContent: 'space-between', alignItems: 'center' }}>
+                        <span style={{ flex: 1, overflow: 'hidden', textOverflow: 'ellipsis' }}>{dir}</span>
                        <X
                          size={14}
-                          style={{
-                            marginLeft: 8,
-                            cursor: 'pointer',
-                            color: '#999'
-                          }}
+                          style={{ marginLeft: 8, cursor: 'pointer', color: '#999' }}
                          onClick={(e) => handleRemoveDirectory(dir, e)}
                        />
                      </div>
@@ -479,14 +429,7 @@ const CodeToolsPage: FC = () => {
                rows={2}
                style={{ fontFamily: 'monospace' }}
              />
-              <div
-                style={{
-                  fontSize: 12,
-                  color: 'var(--color-text-3)',
-                  marginTop: 4
-                }}>
-                {t('code.env_vars_help')}
-              </div>
+              <div style={{ fontSize: 12, color: 'var(--color-text-3)', marginTop: 4 }}>{t('code.env_vars_help')}</div>
            </SettingsItem>

            {/* 终端选择 (macOS 和 Windows) */}
@@ -521,12 +464,7 @@ const CodeToolsPage: FC = () => {
                  selectedTerminal !== terminalApps.cmd &&
                  selectedTerminal !== terminalApps.powershell &&
                  selectedTerminal !== terminalApps.windowsTerminal && (
-                    <div
-                      style={{
-                        fontSize: 12,
-                        color: 'var(--color-text-3)',
-                        marginTop: 4
-                      }}>
+                    <div style={{ fontSize: 12, color: 'var(--color-text-3)', marginTop: 4 }}>
                      {terminalCustomPaths[selectedTerminal]
                        ? `${t('code.custom_path')}: ${terminalCustomPaths[selectedTerminal]}`
                        : t('code.custom_path_required')}
--- a/src/renderer/src/pages/code/index.ts
+++ b/src/renderer/src/pages/code/index.ts
@@ -20,8 +20,7 @@ export const CLI_TOOLS = [
  { value: codeTools.qwenCode, label: 'Qwen Code' },
  { value: codeTools.geminiCli, label: 'Gemini CLI' },
  { value: codeTools.openaiCodex, label: 'OpenAI Codex' },
-  { value: codeTools.iFlowCli, label: 'iFlow CLI' },
-  { value: codeTools.githubCopilotCli, label: 'GitHub Copilot CLI' }
+  { value: codeTools.iFlowCli, label: 'iFlow CLI' }
 ]

 export const GEMINI_SUPPORTED_PROVIDERS = ['aihubmix', 'dmxapi', 'new-api', 'cherryin']
@@ -44,8 +43,7 @@ export const CLI_TOOL_PROVIDER_MAP: Record<string, (providers: Provider[]) => Pr
  [codeTools.qwenCode]: (providers) => providers.filter((p) => p.type.includes('openai')),
  [codeTools.openaiCodex]: (providers) =>
    providers.filter((p) => p.id === 'openai' || OPENAI_CODEX_SUPPORTED_PROVIDERS.includes(p.id)),
-  [codeTools.iFlowCli]: (providers) => providers.filter((p) => p.type.includes('openai')),
-  [codeTools.githubCopilotCli]: () => []
+  [codeTools.iFlowCli]: (providers) => providers.filter((p) => p.type.includes('openai'))
 }

 export const getCodeToolsApiBaseUrl = (model: Model, type: EndpointType) => {
@@ -160,10 +158,6 @@ export const generateToolEnvironment = ({
      env.IFLOW_BASE_URL = baseUrl
      env.IFLOW_MODEL_NAME = model.id
      break
-
-    case codeTools.githubCopilotCli:
-      env.GITHUB_TOKEN = apiKey || ''
-      break
  }

  return env
--- a/src/renderer/src/pages/home/Inputbar/MentionModelsButton.tsx
+++ b/src/renderer/src/pages/home/Inputbar/MentionModelsButton.tsx
@@ -250,19 +250,21 @@ const MentionModelsButton: FC<Props> = ({
          // ESC关闭时的处理：删除 @ 和搜索文本
          if (action === 'esc') {
            // 只有在输入触发且有模型选择动作时才删除@字符和搜索文本
-            const triggerInfo = ctx?.triggerInfo ?? triggerInfoRef.current
-            if (hasModelActionRef.current && triggerInfo?.type === 'input' && triggerInfo?.position !== undefined) {
+            if (
+              hasModelActionRef.current &&
+              ctx.triggerInfo?.type === 'input' &&
+              ctx.triggerInfo?.position !== undefined
+            ) {
              // 基于当前光标 + 搜索词精确定位并删除，position 仅作兜底
              setText((currentText) => {
                const textArea = document.querySelector('.inputbar textarea') as HTMLTextAreaElement | null
                const caret = textArea ? (textArea.selectionStart ?? currentText.length) : currentText.length
-                return removeAtSymbolAndText(currentText, caret, searchText || '', triggerInfo.position!)
+                return removeAtSymbolAndText(currentText, caret, searchText || '', ctx.triggerInfo?.position!)
              })
            }
          }
          // Backspace删除@的情况（delete-symbol）：
          // @ 已经被Backspace自然删除，面板关闭，不需要额外操作
-          triggerInfoRef.current = undefined
        }
      })
    },
--- a/src/renderer/src/pages/home/Markdown/Markdown.tsx
+++ b/src/renderer/src/pages/home/Markdown/Markdown.tsx
@@ -12,6 +12,7 @@ import { removeSvgEmptyLines } from '@renderer/utils/formats'
 import { processLatexBrackets } from '@renderer/utils/markdown'
 import { isEmpty } from 'lodash'
 import { type FC, memo, useCallback, useEffect, useMemo, useRef, useState } from 'react'
+import React from 'react'
 import { useTranslation } from 'react-i18next'
 import ReactMarkdown, { type Components, defaultUrlTransform } from 'react-markdown'
 import rehypeKatex from 'rehype-katex'
@@ -64,6 +65,8 @@ const Markdown: FC<Props> = ({ block, postProcess }) => {
    initialText: block.content
  })

+  const isStreaming = block.status === 'streaming'
+
  useEffect(() => {
    const newContent = block.content || ''
    const oldContent = prevContentRef.current || ''
@@ -85,9 +88,8 @@ const Markdown: FC<Props> = ({ block, postProcess }) => {
    prevBlockIdRef.current = block.id

    // 更新 stream 状态
-    const isStreaming = block.status === 'streaming'
    setIsStreamDone(!isStreaming)
-  }, [block.content, block.id, block.status, addChunk, reset])
+  }, [block.content, block.id, block.status, addChunk, reset, isStreaming])

  const remarkPlugins = useMemo(() => {
    const plugins = [
@@ -130,14 +132,16 @@ const Markdown: FC<Props> = ({ block, postProcess }) => {
      table: (props: any) => <Table {...props} blockId={block.id} />,
      img: (props: any) => <ImageViewer style={{ maxWidth: 500, maxHeight: 500 }} {...props} />,
      pre: (props: any) => <pre style={{ overflow: 'visible' }} {...props} />,
-      p: (props) => {
+      p: SmoothFade((props) => {
        const hasImage = props?.node?.children?.some((child: any) => child.tagName === 'img')
        if (hasImage) return <div {...props} />
        return <p {...props} />
-      },
-      svg: MarkdownSvgRenderer
+      }, isStreaming),
+      svg: MarkdownSvgRenderer,
+      li: SmoothFade((props) => <li {...props} />, isStreaming),
+      span: SmoothFade((props) => <span {...props} />, isStreaming)
    } as Partial<Components>
-  }, [block.id])
+  }, [block.id, isStreaming])

  if (/<style\b[^>]*>/i.test(messageContent)) {
    components.style = MarkdownShadowDOMRenderer as any
@@ -168,3 +172,23 @@ const Markdown: FC<Props> = ({ block, postProcess }) => {
 }

 export default memo(Markdown)
+
+const SmoothFade = (Comp: React.ElementType, isStreaming: boolean) => {
+  const handleAnimationEnd = (e: React.AnimationEvent) => {
+    // 动画结束后移除类名
+    if (e.animationName === 'fadeInWithBlur') {
+      e.currentTarget.classList.remove('animate-[fadeInWithBlur_500ms_ease-out_forwards]')
+      e.currentTarget.classList.remove('opacity-0')
+    }
+  }
+  return ({ children, ...rest }) => {
+    return (
+      <Comp
+        {...rest}
+        className={isStreaming ? 'animate-[fadeInWithBlur_500ms_ease-out_forwards] opacity-0' : ''}
+        onAnimationEnd={handleAnimationEnd}>
+        {children}
+      </Comp>
+    )
+  }
+}
--- a/src/renderer/src/pages/home/Messages/Blocks/ErrorBlock.tsx
+++ b/src/renderer/src/pages/home/Messages/Blocks/ErrorBlock.tsx
@@ -103,8 +103,7 @@ const MessageErrorInfo: React.FC<{ block: ErrorMessageBlock; message: Message }>
  const [showDetailModal, setShowDetailModal] = useState(false)
  const { t } = useTranslation()

-  const onRemoveBlock = (e: React.MouseEvent) => {
-    e.stopPropagation()
+  const onRemoveBlock = () => {
    setTimeoutTimer('onRemoveBlock', () => dispatch(removeBlocksThunk(message.topicId, message.id, [block.id])), 350)
  }

--- a/src/renderer/src/pages/home/Messages/Blocks/PlaceholderBlock.tsx
+++ b/src/renderer/src/pages/home/Messages/Blocks/PlaceholderBlock.tsx
@@ -1,27 +1,20 @@
-import { Spinner } from '@heroui/react'
-import { MessageBlockStatus, MessageBlockType, type PlaceholderMessageBlock } from '@renderer/types/newMessage'
+import { MessageBlockStatus, MessageBlockType } from '@renderer/types/newMessage'
+import { Loader } from '@renderer/ui/loader'
 import React from 'react'
-import styled from 'styled-components'

 interface PlaceholderBlockProps {
-  block: PlaceholderMessageBlock
+  status: MessageBlockStatus
+  type: MessageBlockType
 }
-const PlaceholderBlock: React.FC<PlaceholderBlockProps> = ({ block }) => {
-  if (block.status === MessageBlockStatus.PROCESSING && block.type === MessageBlockType.UNKNOWN) {
+const PlaceholderBlock: React.FC<PlaceholderBlockProps> = ({ status, type }) => {
+  if (status === MessageBlockStatus.PROCESSING && type === MessageBlockType.UNKNOWN) {
    return (
-      <MessageContentLoading>
-        <Spinner color="current" variant="dots" />
-      </MessageContentLoading>
+      <div className="-mt-2">
+        <Loader variant="terminal" />
+      </div>
    )
  }
  return null
 }
-const MessageContentLoading = styled.div`
-  display: flex;
-  flex-direction: row;
-  align-items: center;
-  height: 32px;
-  margin-top: -5px;
-  margin-bottom: 5px;
-`
+
 export default React.memo(PlaceholderBlock)
--- a/src/renderer/src/pages/home/Messages/Blocks/ThinkingBlock.tsx
+++ b/src/renderer/src/pages/home/Messages/Blocks/ThinkingBlock.tsx
@@ -5,7 +5,7 @@ import { useSettings } from '@renderer/hooks/useSettings'
 import { useTemporaryValue } from '@renderer/hooks/useTemporaryValue'
 import { MessageBlockStatus, type ThinkingMessageBlock } from '@renderer/types/newMessage'
 import { Collapse, message as antdMessage, Tooltip } from 'antd'
-import { memo, useCallback, useEffect, useMemo, useRef, useState } from 'react'
+import { memo, useCallback, useEffect, useMemo, useState } from 'react'
 import { useTranslation } from 'react-i18next'
 import styled from 'styled-components'

@@ -105,37 +105,30 @@ const ThinkingBlock: React.FC<Props> = ({ block }) => {
 const ThinkingTimeSeconds = memo(
  ({ blockThinkingTime, isThinking }: { blockThinkingTime: number; isThinking: boolean }) => {
    const { t } = useTranslation()
-    const [displayTime, setDisplayTime] = useState(blockThinkingTime)
+    // const [thinkingTime, setThinkingTime] = useState(blockThinkingTime || 0)

-    const timer = useRef<NodeJS.Timeout | null>(null)
+    // FIXME: 这里统计的和请求处统计的有一定误差
+    // useEffect(() => {
+    //   let timer: NodeJS.Timeout | null = null
+    //   if (isThinking) {
+    //     timer = setInterval(() => {
+    //       setThinkingTime((prev) => prev + 100)
+    //     }, 100)
+    //   } else if (timer) {
+    //     // 立即清除计时器
+    //     clearInterval(timer)
+    //     timer = null
+    //   }

-    useEffect(() => {
-      if (isThinking) {
-        if (!timer.current) {
-          timer.current = setInterval(() => {
-            setDisplayTime((prev) => prev + 100)
-          }, 100)
-        }
-      } else {
-        if (timer.current) {
-          clearInterval(timer.current)
-          timer.current = null
-        }
-        setDisplayTime(blockThinkingTime)
-      }
+    //   return () => {
+    //     if (timer) {
+    //       clearInterval(timer)
+    //       timer = null
+    //     }
+    //   }
+    // }, [isThinking])

-      return () => {
-        if (timer.current) {
-          clearInterval(timer.current)
-          timer.current = null
-        }
-      }
-    }, [isThinking, blockThinkingTime])
-
-    const thinkingTimeSeconds = useMemo(
-      () => ((displayTime < 1000 ? 100 : displayTime) / 1000).toFixed(1),
-      [displayTime]
-    )
+    const thinkingTimeSeconds = useMemo(() => (blockThinkingTime / 1000).toFixed(1), [blockThinkingTime])

    return isThinking
      ? t('chat.thinking', {
--- a/src/renderer/src/pages/home/Messages/Blocks/tests/ThinkingBlock.test.tsx
+++ b/src/renderer/src/pages/home/Messages/Blocks/tests/ThinkingBlock.test.tsx
@@ -235,12 +235,13 @@ describe('ThinkingBlock', () => {
      renderThinkingBlock(thinkingBlock)

      const activeTimeText = getThinkingTimeText()
+      expect(activeTimeText).toHaveTextContent('1.0s')
      expect(activeTimeText).toHaveTextContent('Thinking...')
    })

    it('should handle extreme thinking times correctly', () => {
      const testCases = [
-        { thinking_millsec: 0, expectedTime: '0.1s' }, // New logic: values < 1000ms display as 0.1s
+        { thinking_millsec: 0, expectedTime: '0.0s' },
        { thinking_millsec: 86400000, expectedTime: '86400.0s' }, // 1 day
        { thinking_millsec: 259200000, expectedTime: '259200.0s' } // 3 days
      ]
--- a/src/renderer/src/pages/home/Messages/Blocks/index.tsx
+++ b/src/renderer/src/pages/home/Messages/Blocks/index.tsx
@@ -215,15 +215,7 @@ const MessageBlockRenderer: React.FC<Props> = ({ blocks, message }) => {
      })}
      {isProcessing && (
        <AnimatedBlockWrapper key="message-loading-placeholder" enableAnimation={true}>
-          <PlaceholderBlock
-            block={{
-              id: `loading-${message.id}`,
-              messageId: message.id,
-              type: MessageBlockType.UNKNOWN,
-              status: MessageBlockStatus.PROCESSING,
-              createdAt: new Date().toISOString()
-            }}
-          />
+          <PlaceholderBlock type={MessageBlockType.UNKNOWN} status={MessageBlockStatus.PROCESSING} />
        </AnimatedBlockWrapper>
      )}
    </AnimatePresence>
--- a/src/renderer/src/pages/home/Messages/Message.tsx
+++ b/src/renderer/src/pages/home/Messages/Message.tsx
@@ -1,6 +1,4 @@
-import { cn } from '@heroui/react'
 import { loggerService } from '@logger'
-import HorizontalScrollContainer from '@renderer/components/HorizontalScrollContainer'
 import Scrollbar from '@renderer/components/Scrollbar'
 import { useMessageEditing } from '@renderer/context/MessageEditingContext'
 import { useAssistant } from '@renderer/hooks/useAssistant'
@@ -227,28 +225,20 @@ const MessageItem: FC<Props> = ({
              </MessageErrorBoundary>
            </MessageContentContainer>
            {showMenubar && (
-              <MessageFooter className="MessageFooter">
-                <HorizontalScrollContainer
-                  classNames={{
-                    content: cn(
-                      'flex-1 items-center justify-between',
-                      isLastMessage && messageStyle === 'plain' ? 'flex-row-reverse' : 'flex-row'
-                    )
-                  }}>
-                  <MessageMenubar
-                    message={message}
-                    assistant={assistant}
-                    model={model}
-                    index={index}
-                    topic={topic}
-                    isLastMessage={isLastMessage}
-                    isAssistantMessage={isAssistantMessage}
-                    isGrouped={isGrouped}
-                    messageContainerRef={messageContainerRef as React.RefObject<HTMLDivElement>}
-                    setModel={setModel}
-                    onUpdateUseful={onUpdateUseful}
-                  />
-                </HorizontalScrollContainer>
+              <MessageFooter className="MessageFooter" $isLastMessage={isLastMessage} $messageStyle={messageStyle}>
+                <MessageMenubar
+                  message={message}
+                  assistant={assistant}
+                  model={model}
+                  index={index}
+                  topic={topic}
+                  isLastMessage={isLastMessage}
+                  isAssistantMessage={isAssistantMessage}
+                  isGrouped={isGrouped}
+                  messageContainerRef={messageContainerRef as React.RefObject<HTMLDivElement>}
+                  setModel={setModel}
+                  onUpdateUseful={onUpdateUseful}
+                />
              </MessageFooter>
            )}
          </>
@@ -292,8 +282,10 @@ const MessageContentContainer = styled(Scrollbar)`
  overflow-y: auto;
 `

-const MessageFooter = styled.div`
+const MessageFooter = styled.div<{ $isLastMessage: boolean; $messageStyle: 'plain' | 'bubble' }>`
  display: flex;
+  flex-direction: ${({ $isLastMessage, $messageStyle }) =>
+    $isLastMessage && $messageStyle === 'plain' ? 'row-reverse' : 'row'};
  align-items: center;
  justify-content: space-between;
  gap: 10px;
--- a/src/renderer/src/pages/home/Messages/MessageGroup.tsx
+++ b/src/renderer/src/pages/home/Messages/MessageGroup.tsx
@@ -337,29 +337,17 @@ const GroupContainer = styled.div`
 const GridContainer = styled(Scrollbar)<{ $count: number; $gridColumns: number }>`
  width: 100%;
  display: grid;
+  overflow-y: visible;
  gap: 16px;
-
  &.horizontal {
    padding-bottom: 4px;
    grid-template-columns: repeat(${({ $count }) => $count}, minmax(420px, 1fr));
-    overflow-y: hidden;
    overflow-x: auto;
-    &::-webkit-scrollbar {
-      height: 6px;
-    }
-    &::-webkit-scrollbar-thumb {
-      background: var(--color-scrollbar-thumb);
-      border-radius: var(--scrollbar-thumb-radius);
-    }
-    &::-webkit-scrollbar-thumb:hover {
-      background: var(--color-scrollbar-thumb-hover);
-    }
  }
  &.fold,
  &.vertical {
    grid-template-columns: repeat(1, minmax(0, 1fr));
    gap: 8px;
-    overflow: hidden;
  }
  &.grid {
    grid-template-columns: repeat(
@@ -367,15 +355,11 @@ const GridContainer = styled(Scrollbar)<{ $count: number; $gridColumns: number }
      minmax(0, 1fr)
    );
    grid-template-rows: auto;
-    overflow-y: auto;
-    overflow-x: hidden;
  }

  &.multi-select-mode {
    grid-template-columns: repeat(1, minmax(0, 1fr));
    gap: 10px;
-    overflow-y: auto;
-    overflow-x: hidden;
    .grid {
      height: auto;
    }
@@ -401,7 +385,7 @@ interface MessageWrapperProps {
 const MessageWrapper = styled.div<MessageWrapperProps>`
  &.horizontal {
    padding: 1px;
-    /* overflow-y: auto; */
+    overflow-y: auto;
    .message {
      height: 100%;
      border: 0.5px solid var(--color-border);
@@ -421,9 +405,8 @@ const MessageWrapper = styled.div<MessageWrapperProps>`
    }
  }
  &.grid {
-    display: block;
    height: 300px;
-    overflow: hidden;
+    overflow-y: hidden;
    border: 0.5px solid var(--color-border);
    border-radius: 10px;
    cursor: pointer;
--- a/src/renderer/src/pages/home/Messages/Tools/MessageKnowledgeSearch.tsx
+++ b/src/renderer/src/pages/home/Messages/Tools/MessageKnowledgeSearch.tsx
@@ -1,13 +1,13 @@
 import { KnowledgeSearchToolInput, KnowledgeSearchToolOutput } from '@renderer/aiCore/tools/KnowledgeSearchTool'
 import Spinner from '@renderer/components/Spinner'
 import i18n from '@renderer/i18n'
-import { NormalToolResponse } from '@renderer/types'
+import { MCPToolResponse } from '@renderer/types'
 import { Typography } from 'antd'
 import { FileSearch } from 'lucide-react'
 import styled from 'styled-components'

 const { Text } = Typography
-export function MessageKnowledgeSearchToolTitle({ toolResponse }: { toolResponse: NormalToolResponse }) {
+export function MessageKnowledgeSearchToolTitle({ toolResponse }: { toolResponse: MCPToolResponse }) {
  const toolInput = toolResponse.arguments as KnowledgeSearchToolInput
  const toolOutput = toolResponse.response as KnowledgeSearchToolOutput

@@ -28,7 +28,7 @@ export function MessageKnowledgeSearchToolTitle({ toolResponse }: { toolResponse
  )
 }

-export function MessageKnowledgeSearchToolBody({ toolResponse }: { toolResponse: NormalToolResponse }) {
+export function MessageKnowledgeSearchToolBody({ toolResponse }: { toolResponse: MCPToolResponse }) {
  const toolOutput = toolResponse.response as KnowledgeSearchToolOutput

  return toolResponse.status === 'done' ? (
--- a/src/renderer/src/pages/home/Messages/Tools/MessageMcpTool.tsx
+++ b/src/renderer/src/pages/home/Messages/Tools/MessageMcpTool.tsx
@@ -4,7 +4,6 @@ import { useCodeStyle } from '@renderer/context/CodeStyleProvider'
 import { useMCPServers } from '@renderer/hooks/useMCPServers'
 import { useSettings } from '@renderer/hooks/useSettings'
 import { useTimer } from '@renderer/hooks/useTimer'
-import { MCPToolResponse } from '@renderer/types'
 import type { ToolMessageBlock } from '@renderer/types/newMessage'
 import { isToolAutoApproved } from '@renderer/utils/mcp-tools'
 import { cancelToolAction, confirmToolAction } from '@renderer/utils/userConfirmation'
@@ -58,7 +57,7 @@ const MessageMcpTool: FC<Props> = ({ block }) => {
  const [progress, setProgress] = useState<number>(0)
  const { setTimeoutTimer } = useTimer()

-  const toolResponse = block.metadata?.rawMcpToolResponse as MCPToolResponse
+  const toolResponse = block.metadata?.rawMcpToolResponse

  const { id, tool, status, response } = toolResponse!
  const isPending = status === 'pending'
--- a/src/renderer/src/pages/home/Messages/Tools/MessageMemorySearch.tsx
+++ b/src/renderer/src/pages/home/Messages/Tools/MessageMemorySearch.tsx
@@ -1,6 +1,6 @@
 import { MemorySearchToolInput, MemorySearchToolOutput } from '@renderer/aiCore/tools/MemorySearchTool'
 import Spinner from '@renderer/components/Spinner'
-import { NormalToolResponse } from '@renderer/types'
+import { MCPToolResponse } from '@renderer/types'
 import { Typography } from 'antd'
 import { ChevronRight } from 'lucide-react'
 import { useTranslation } from 'react-i18next'
@@ -8,7 +8,7 @@ import styled from 'styled-components'

 const { Text } = Typography

-export const MessageMemorySearchToolTitle = ({ toolResponse }: { toolResponse: NormalToolResponse }) => {
+export const MessageMemorySearchToolTitle = ({ toolResponse }: { toolResponse: MCPToolResponse }) => {
  const { t } = useTranslation()
  const toolInput = toolResponse.arguments as MemorySearchToolInput
  const toolOutput = toolResponse.response as MemorySearchToolOutput
--- a/src/renderer/src/pages/home/Messages/Tools/MessageTool.tsx
+++ b/src/renderer/src/pages/home/Messages/Tools/MessageTool.tsx
@@ -1,4 +1,4 @@
-import { NormalToolResponse } from '@renderer/types'
+import { MCPToolResponse } from '@renderer/types'
 import type { ToolMessageBlock } from '@renderer/types/newMessage'
 import { Collapse } from 'antd'

@@ -11,9 +11,8 @@ interface Props {
 }
 const prefix = 'builtin_'

-const ChooseTool = (toolResponse: NormalToolResponse): { label: React.ReactNode; body: React.ReactNode } | null => {
+const ChooseTool = (toolResponse: MCPToolResponse): { label: React.ReactNode; body: React.ReactNode } | null => {
  let toolName = toolResponse.tool.name
-  const toolType = toolResponse.tool.type
  if (toolName.startsWith(prefix)) {
    toolName = toolName.slice(prefix.length)
  }
@@ -21,12 +20,10 @@ const ChooseTool = (toolResponse: NormalToolResponse): { label: React.ReactNode;
  switch (toolName) {
    case 'web_search':
    case 'web_search_preview':
-      return toolType === 'provider'
-        ? null
-        : {
-            label: <MessageWebSearchToolTitle toolResponse={toolResponse} />,
-            body: null
-          }
+      return {
+        label: <MessageWebSearchToolTitle toolResponse={toolResponse} />,
+        body: null
+      }
    case 'knowledge_search':
      return {
        label: <MessageKnowledgeSearchToolTitle toolResponse={toolResponse} />,
@@ -44,7 +41,7 @@ const ChooseTool = (toolResponse: NormalToolResponse): { label: React.ReactNode;

 export default function MessageTool({ block }: Props) {
  // FIXME: 语义错误，这里已经不是 MCP tool 了,更改rawMcpToolResponse需要改用户数据, 所以暂时保留
-  const toolResponse = block.metadata?.rawMcpToolResponse as NormalToolResponse
+  const toolResponse = block.metadata?.rawMcpToolResponse

  if (!toolResponse) return null

--- a/src/renderer/src/pages/home/Messages/Tools/MessageWebSearch.tsx
+++ b/src/renderer/src/pages/home/Messages/Tools/MessageWebSearch.tsx
@@ -1,6 +1,6 @@
 import { WebSearchToolInput, WebSearchToolOutput } from '@renderer/aiCore/tools/WebSearchTool'
 import Spinner from '@renderer/components/Spinner'
-import { NormalToolResponse } from '@renderer/types'
+import { MCPToolResponse } from '@renderer/types'
 import { Typography } from 'antd'
 import { Search } from 'lucide-react'
 import { useTranslation } from 'react-i18next'
@@ -8,7 +8,7 @@ import styled from 'styled-components'

 const { Text } = Typography

-export const MessageWebSearchToolTitle = ({ toolResponse }: { toolResponse: NormalToolResponse }) => {
+export const MessageWebSearchToolTitle = ({ toolResponse }: { toolResponse: MCPToolResponse }) => {
  const { t } = useTranslation()
  const toolInput = toolResponse.arguments as WebSearchToolInput
  const toolOutput = toolResponse.response as WebSearchToolOutput
--- a/src/renderer/src/pages/home/Tabs/TopicsTab.tsx
+++ b/src/renderer/src/pages/home/Tabs/TopicsTab.tsx
@@ -525,11 +525,6 @@ const Topics: FC<Props> = ({ assistant: _assistant, activeTopic, setActiveTopic,
              onContextMenu={() => setTargetTopic(topic)}
              className={classNames(isActive ? 'active' : '', singlealone ? 'singlealone' : '')}
              onClick={editingTopicId === topic.id && topicEdit.isEditing ? undefined : () => onSwitchTopic(topic)}
-              onDoubleClick={() => {
-                if (editingTopicId === topic.id && topicEdit.isEditing) return
-                setEditingTopicId(topic.id)
-                topicEdit.startEdit(topic.name)
-              }}
              style={{
                borderRadius,
                cursor: editingTopicId === topic.id && topicEdit.isEditing ? 'default' : 'pointer'
@@ -546,7 +541,13 @@ const Topics: FC<Props> = ({ assistant: _assistant, activeTopic, setActiveTopic,
                    onClick={(e) => e.stopPropagation()}
                  />
                ) : (
-                  <TopicName className={getTopicNameClassName()} title={topicName}>
+                  <TopicName
+                    className={getTopicNameClassName()}
+                    title={topicName}
+                    onDoubleClick={() => {
+                      setEditingTopicId(topic.id)
+                      topicEdit.startEdit(topic.name)
+                    }}>
                    {topicName}
                  </TopicName>
                )}
@@ -570,8 +571,7 @@ const Topics: FC<Props> = ({ assistant: _assistant, activeTopic, setActiveTopic,
                        } else {
                          handleDeleteClick(topic.id, e)
                        }
-                      }}
-                      onDoubleClick={(e) => e.stopPropagation()}>
+                      }}>
                      {deletingTopicId === topic.id ? (
                        <DeleteIcon size={14} color="var(--color-error)" style={{ pointerEvents: 'none' }} />
                      ) : (
--- a/src/renderer/src/pages/home/Tabs/components/AssistantItem.tsx
+++ b/src/renderer/src/pages/home/Tabs/components/AssistantItem.tsx
@@ -386,7 +386,6 @@ const Container = styled.div`
  border-radius: var(--list-item-border-radius);
  border: 0.5px solid transparent;
  width: calc(var(--assistants-width) - 20px);
-  cursor: pointer;

  &:hover {
    background-color: var(--color-list-item-hover);
--- a/src/renderer/src/pages/home/components/UpdateAppButton.tsx
+++ b/src/renderer/src/pages/home/components/UpdateAppButton.tsx
@@ -1,6 +1,4 @@
 import { SyncOutlined } from '@ant-design/icons'
-import { useDisclosure } from '@heroui/react'
-import UpdateDialog from '@renderer/components/UpdateDialog'
 import { useRuntime } from '@renderer/hooks/useRuntime'
 import { useSettings } from '@renderer/hooks/useSettings'
 import { Button } from 'antd'
@@ -12,7 +10,6 @@ const UpdateAppButton: FC = () => {
  const { update } = useRuntime()
  const { autoCheckUpdate } = useSettings()
  const { t } = useTranslation()
-  const { isOpen, onOpen, onClose } = useDisclosure()

  if (!update) {
    return null
@@ -26,15 +23,13 @@ const UpdateAppButton: FC = () => {
    <Container>
      <UpdateButton
        className="nodrag"
-        onClick={onOpen}
+        onClick={() => window.api.showUpdateDialog()}
        icon={<SyncOutlined />}
        color="orange"
        variant="outlined"
        size="small">
        {t('button.update_available')}
      </UpdateButton>
-
-      <UpdateDialog isOpen={isOpen} onClose={onClose} releaseInfo={update.info || null} />
    </Container>
  )
 }
--- a/src/renderer/src/pages/minapps/MinAppPage.tsx
+++ b/src/renderer/src/pages/minapps/MinAppPage.tsx
@@ -14,7 +14,6 @@ import styled from 'styled-components'

 // Tab 模式下新的页面壳，不再直接创建 WebView，而是依赖全局 MinAppTabsPool
 import MinimalToolbar from './components/MinimalToolbar'
-import WebviewSearch from './components/WebviewSearch'

 const logger = loggerService.withContext('MinAppPage')

@@ -185,7 +184,6 @@ const MinAppPage: FC = () => {
          onOpenDevTools={handleOpenDevTools}
        />
      </ToolbarWrapper>
-      <WebviewSearch webviewRef={webviewRef} isWebviewReady={isReady} appId={app.id} />
      {!isReady && (
        <LoadingMask>
          <Avatar src={app.logo} size={60} style={{ border: '1px solid var(--color-border)' }} />
--- a/src/renderer/src/pages/minapps/components/WebviewSearch.tsx
+++ b/src/renderer/src/pages/minapps/components/WebviewSearch.tsx
@@ -1,298 +0,0 @@
-import { Button, Input } from '@heroui/react'
-import { loggerService } from '@logger'
-import type { WebviewTag } from 'electron'
-import { ChevronDown, ChevronUp, X } from 'lucide-react'
-import { FC, useCallback, useEffect, useRef, useState } from 'react'
-import { useTranslation } from 'react-i18next'
-
-type FoundInPageResult = Electron.FoundInPageResult
-
-interface WebviewSearchProps {
-  webviewRef: React.RefObject<WebviewTag | null>
-  isWebviewReady: boolean
-  appId: string
-}
-
-const logger = loggerService.withContext('WebviewSearch')
-
-const WebviewSearch: FC<WebviewSearchProps> = ({ webviewRef, isWebviewReady, appId }) => {
-  const { t } = useTranslation()
-  const [isVisible, setIsVisible] = useState(false)
-  const [query, setQuery] = useState('')
-  const [matchCount, setMatchCount] = useState(0)
-  const [activeIndex, setActiveIndex] = useState(0)
-  const [currentWebview, setCurrentWebview] = useState<WebviewTag | null>(null)
-  const inputRef = useRef<HTMLInputElement>(null)
-  const focusFrameRef = useRef<number | null>(null)
-  const lastAppIdRef = useRef<string>(appId)
-  const attachedWebviewRef = useRef<WebviewTag | null>(null)
-
-  const focusInput = useCallback(() => {
-    if (focusFrameRef.current !== null) {
-      window.cancelAnimationFrame(focusFrameRef.current)
-      focusFrameRef.current = null
-    }
-    focusFrameRef.current = window.requestAnimationFrame(() => {
-      inputRef.current?.focus()
-      inputRef.current?.select()
-    })
-  }, [])
-
-  const resetSearchState = useCallback((options?: { keepQuery?: boolean }) => {
-    if (!options?.keepQuery) {
-      setQuery('')
-    }
-    setMatchCount(0)
-    setActiveIndex(0)
-  }, [])
-
-  const stopSearch = useCallback(() => {
-    const target = webviewRef.current ?? attachedWebviewRef.current
-    if (!target) return
-    try {
-      target.stopFindInPage('clearSelection')
-    } catch (error) {
-      logger.error('stopFindInPage failed', { error })
-    }
-  }, [webviewRef])
-
-  const closeSearch = useCallback(() => {
-    setIsVisible(false)
-    stopSearch()
-    resetSearchState({ keepQuery: true })
-  }, [resetSearchState, stopSearch])
-
-  const performSearch = useCallback(
-    (text: string, options?: Electron.FindInPageOptions) => {
-      const target = webviewRef.current ?? attachedWebviewRef.current
-      if (!target) {
-        logger.debug('Skip performSearch: webview not attached')
-        return
-      }
-      if (!text) {
-        stopSearch()
-        resetSearchState({ keepQuery: true })
-        return
-      }
-      try {
-        target.findInPage(text, options)
-      } catch (error) {
-        logger.error('findInPage failed', { error })
-        window.toast?.error(t('common.error'))
-      }
-    },
-    [resetSearchState, stopSearch, t, webviewRef]
-  )
-
-  const handleFoundInPage = useCallback((event: Event & { result?: FoundInPageResult }) => {
-    if (!event.result) return
-
-    const { activeMatchOrdinal, matches } = event.result
-
-    if (matches !== undefined) {
-      setMatchCount(matches)
-    }
-
-    if (activeMatchOrdinal !== undefined) {
-      setActiveIndex(activeMatchOrdinal)
-    }
-  }, [])
-
-  const openSearch = useCallback(() => {
-    if (!isWebviewReady) {
-      logger.debug('Skip openSearch: webview not ready')
-      return
-    }
-    setIsVisible(true)
-    focusInput()
-  }, [focusInput, isWebviewReady])
-
-  const goToNext = useCallback(() => {
-    if (!query) return
-    performSearch(query, { forward: true, findNext: true })
-  }, [performSearch, query])
-
-  const goToPrevious = useCallback(() => {
-    if (!query) return
-    performSearch(query, { forward: false, findNext: true })
-  }, [performSearch, query])
-
-  useEffect(() => {
-    const nextWebview = webviewRef.current ?? null
-    if (currentWebview === nextWebview) return
-    setCurrentWebview(nextWebview)
-  }, [currentWebview, webviewRef])
-
-  useEffect(() => {
-    const target = currentWebview
-    if (!target) {
-      attachedWebviewRef.current = null
-      return
-    }
-
-    const handle = handleFoundInPage
-    attachedWebviewRef.current = target
-    target.addEventListener('found-in-page', handle)
-
-    return () => {
-      target.removeEventListener('found-in-page', handle)
-      if (attachedWebviewRef.current === target) {
-        try {
-          target.stopFindInPage('clearSelection')
-        } catch (error) {
-          logger.error('stopFindInPage failed', { error })
-        }
-        attachedWebviewRef.current = null
-      }
-    }
-  }, [currentWebview, handleFoundInPage])
-
-  useEffect(() => {
-    if (!isVisible) return
-    focusInput()
-  }, [focusInput, isVisible])
-
-  useEffect(() => {
-    if (!isVisible) return
-    if (!query) {
-      performSearch('')
-      return
-    }
-    performSearch(query)
-  }, [currentWebview, isVisible, performSearch, query])
-
-  useEffect(() => {
-    const handleKeydown = (event: KeyboardEvent) => {
-      if ((event.ctrlKey || event.metaKey) && event.key.toLowerCase() === 'f') {
-        event.preventDefault()
-        openSearch()
-        return
-      }
-
-      if (!isVisible) return
-
-      if (event.key === 'Escape') {
-        event.preventDefault()
-        closeSearch()
-        return
-      }
-
-      if (event.key === 'Enter') {
-        event.preventDefault()
-        if (event.shiftKey) {
-          goToPrevious()
-        } else {
-          goToNext()
-        }
-      }
-    }
-
-    window.addEventListener('keydown', handleKeydown, true)
-    return () => {
-      window.removeEventListener('keydown', handleKeydown, true)
-    }
-  }, [closeSearch, goToNext, goToPrevious, isVisible, openSearch])
-
-  useEffect(() => {
-    if (!isWebviewReady) {
-      setIsVisible(false)
-      resetSearchState()
-      stopSearch()
-      return
-    }
-  }, [isWebviewReady, resetSearchState, stopSearch])
-
-  useEffect(() => {
-    if (!appId) return
-    if (lastAppIdRef.current === appId) return
-    lastAppIdRef.current = appId
-    setIsVisible(false)
-    resetSearchState()
-    stopSearch()
-  }, [appId, resetSearchState, stopSearch])
-
-  useEffect(() => {
-    return () => {
-      stopSearch()
-      if (focusFrameRef.current !== null) {
-        window.cancelAnimationFrame(focusFrameRef.current)
-        focusFrameRef.current = null
-      }
-    }
-  }, [stopSearch])
-
-  if (!isVisible) {
-    return null
-  }
-
-  const matchLabel = `${matchCount > 0 ? Math.max(activeIndex, 1) : 0}/${matchCount}`
-  const noResultTitle = matchCount === 0 && query ? t('common.no_results') : undefined
-  const disableNavigation = !query || matchCount === 0
-
-  return (
-    <div className="pointer-events-auto absolute top-3 right-3 z-50 flex items-center gap-2 rounded-xl border border-default-200 bg-background px-2 py-1 shadow-lg">
-      <Input
-        ref={inputRef}
-        autoFocus
-        value={query}
-        onValueChange={setQuery}
-        spellCheck={'false'}
-        placeholder={t('common.search')}
-        size="sm"
-        radius="sm"
-        variant="flat"
-        classNames={{
-          base: 'w-[240px]',
-          inputWrapper:
-            'h-8 bg-transparent border border-transparent shadow-none hover:border-transparent hover:bg-transparent focus:border-transparent data-[hover=true]:border-transparent data-[focus=true]:border-transparent data-[focus-visible=true]:outline-none data-[focus-visible=true]:ring-0',
-          input: 'text-small focus:outline-none focus-visible:outline-none',
-          innerWrapper: 'gap-0'
-        }}
-      />
-      <span
-        className="min-w-[44px] text-center text-default-500 text-small tabular-nums"
-        title={noResultTitle}
-        role="status"
-        aria-live="polite"
-        aria-atomic="true">
-        {matchLabel}
-      </span>
-      <div className="h-4 w-px bg-default-200" />
-      <Button
-        size="sm"
-        variant="light"
-        radius="full"
-        isIconOnly
-        onPress={goToPrevious}
-        isDisabled={disableNavigation}
-        aria-label="Previous match"
-        className="text-default-500 hover:text-default-900">
-        <ChevronUp size={16} />
-      </Button>
-      <Button
-        size="sm"
-        variant="light"
-        radius="full"
-        isIconOnly
-        onPress={goToNext}
-        isDisabled={disableNavigation}
-        aria-label="Next match"
-        className="text-default-500 hover:text-default-900">
-        <ChevronDown size={16} />
-      </Button>
-      <div className="h-4 w-px bg-default-200" />
-      <Button
-        size="sm"
-        variant="light"
-        radius="full"
-        isIconOnly
-        onPress={closeSearch}
-        aria-label={t('common.close')}
-        className="text-default-500 hover:text-default-900">
-        <X size={16} />
-      </Button>
-    </div>
-  )
-}
-
-export default WebviewSearch
--- a/src/renderer/src/pages/minapps/components/tests/WebviewSearch.test.tsx
+++ b/src/renderer/src/pages/minapps/components/tests/WebviewSearch.test.tsx
@@ -1,237 +0,0 @@
-import { act, fireEvent, render, screen, waitFor } from '@testing-library/react'
-import userEvent from '@testing-library/user-event'
-import type { WebviewTag } from 'electron'
-import { afterAll, afterEach, beforeAll, beforeEach, describe, expect, it, vi } from 'vitest'
-
-import WebviewSearch from '../WebviewSearch'
-
-const translations: Record<string, string> = {
-  'common.close': 'Close',
-  'common.error': 'Error',
-  'common.no_results': 'No results',
-  'common.search': 'Search'
-}
-
-vi.mock('react-i18next', () => ({
-  useTranslation: () => ({
-    t: (key: string) => translations[key] ?? key
-  })
-}))
-
-const createWebviewMock = () => {
-  const listeners = new Map<string, Set<(event: Event & { result?: Electron.FoundInPageResult }) => void>>()
-  const findInPageMock = vi.fn()
-  const stopFindInPageMock = vi.fn()
-  const webview = {
-    addEventListener: vi.fn(
-      (type: string, listener: (event: Event & { result?: Electron.FoundInPageResult }) => void) => {
-        if (!listeners.has(type)) {
-          listeners.set(type, new Set())
-        }
-        listeners.get(type)!.add(listener)
-      }
-    ),
-    removeEventListener: vi.fn(
-      (type: string, listener: (event: Event & { result?: Electron.FoundInPageResult }) => void) => {
-        listeners.get(type)?.delete(listener)
-      }
-    ),
-    findInPage: findInPageMock as unknown as WebviewTag['findInPage'],
-    stopFindInPage: stopFindInPageMock as unknown as WebviewTag['stopFindInPage']
-  } as unknown as WebviewTag
-
-  const emit = (type: string, result?: Electron.FoundInPageResult) => {
-    listeners.get(type)?.forEach((listener) => {
-      const event = new CustomEvent(type) as Event & { result?: Electron.FoundInPageResult }
-      event.result = result
-      listener(event)
-    })
-  }
-
-  return {
-    emit,
-    findInPageMock,
-    stopFindInPageMock,
-    webview
-  }
-}
-
-const openSearchOverlay = async () => {
-  await act(async () => {
-    window.dispatchEvent(new KeyboardEvent('keydown', { key: 'f', ctrlKey: true }))
-  })
-  await waitFor(() => {
-    expect(screen.getByPlaceholderText('Search')).toBeInTheDocument()
-  })
-}
-
-const originalRAF = window.requestAnimationFrame
-const originalCAF = window.cancelAnimationFrame
-
-const requestAnimationFrameMock = vi.fn((callback: FrameRequestCallback) => {
-  callback(0)
-  return 1
-})
-const cancelAnimationFrameMock = vi.fn()
-
-beforeAll(() => {
-  Object.defineProperty(window, 'requestAnimationFrame', {
-    value: requestAnimationFrameMock,
-    writable: true
-  })
-  Object.defineProperty(window, 'cancelAnimationFrame', {
-    value: cancelAnimationFrameMock,
-    writable: true
-  })
-})
-
-afterAll(() => {
-  Object.defineProperty(window, 'requestAnimationFrame', {
-    value: originalRAF
-  })
-  Object.defineProperty(window, 'cancelAnimationFrame', {
-    value: originalCAF
-  })
-})
-
-describe('WebviewSearch', () => {
-  const toastMock = {
-    error: vi.fn(),
-    success: vi.fn(),
-    warning: vi.fn(),
-    info: vi.fn(),
-    addToast: vi.fn()
-  }
-
-  beforeEach(() => {
-    Object.assign(window, { toast: toastMock })
-  })
-
-  afterEach(() => {
-    vi.clearAllMocks()
-  })
-
-  it('opens the search overlay with keyboard shortcut', async () => {
-    const { webview } = createWebviewMock()
-    const webviewRef = { current: webview } as React.RefObject<WebviewTag | null>
-
-    render(<WebviewSearch webviewRef={webviewRef} isWebviewReady appId="app-1" />)
-
-    expect(screen.queryByPlaceholderText('Search')).not.toBeInTheDocument()
-
-    await openSearchOverlay()
-
-    expect(screen.getByPlaceholderText('Search')).toBeInTheDocument()
-  })
-
-  it('performs searches and navigates between results', async () => {
-    const { emit, findInPageMock, webview } = createWebviewMock()
-    const webviewRef = { current: webview } as React.RefObject<WebviewTag | null>
-    const user = userEvent.setup()
-
-    render(<WebviewSearch webviewRef={webviewRef} isWebviewReady appId="app-1" />)
-    await openSearchOverlay()
-
-    const input = screen.getByRole('textbox')
-    await user.type(input, 'Cherry')
-
-    await waitFor(() => {
-      expect(findInPageMock).toHaveBeenCalledWith('Cherry', undefined)
-    })
-
-    await act(async () => {
-      emit('found-in-page', {
-        requestId: 1,
-        matches: 3,
-        activeMatchOrdinal: 1,
-        selectionArea: undefined as unknown as Electron.Rectangle,
-        finalUpdate: false
-      } as Electron.FoundInPageResult)
-    })
-
-    const nextButton = screen.getByRole('button', { name: 'Next match' })
-    await waitFor(() => {
-      expect(nextButton).not.toBeDisabled()
-    })
-    await user.click(nextButton)
-    await waitFor(() => {
-      expect(findInPageMock).toHaveBeenLastCalledWith('Cherry', { forward: true, findNext: true })
-    })
-
-    const previousButton = screen.getByRole('button', { name: 'Previous match' })
-    await user.click(previousButton)
-    await waitFor(() => {
-      expect(findInPageMock).toHaveBeenLastCalledWith('Cherry', { forward: false, findNext: true })
-    })
-  })
-
-  it('clears search state when appId changes', async () => {
-    const { findInPageMock, stopFindInPageMock, webview } = createWebviewMock()
-    const webviewRef = { current: webview } as React.RefObject<WebviewTag | null>
-    const user = userEvent.setup()
-
-    const { rerender } = render(<WebviewSearch webviewRef={webviewRef} isWebviewReady appId="app-1" />)
-    await openSearchOverlay()
-
-    const input = screen.getByRole('textbox')
-    await user.type(input, 'Cherry')
-    await waitFor(() => {
-      expect(findInPageMock).toHaveBeenCalled()
-    })
-
-    await act(async () => {
-      rerender(<WebviewSearch webviewRef={webviewRef} isWebviewReady appId="app-2" />)
-    })
-
-    await waitFor(() => {
-      expect(screen.queryByPlaceholderText('Search')).not.toBeInTheDocument()
-    })
-    expect(stopFindInPageMock).toHaveBeenCalledWith('clearSelection')
-  })
-
-  it('shows toast error when search fails', async () => {
-    const { findInPageMock, webview } = createWebviewMock()
-    findInPageMock.mockImplementation(() => {
-      throw new Error('findInPage failed')
-    })
-    const webviewRef = { current: webview } as React.RefObject<WebviewTag | null>
-    const user = userEvent.setup()
-
-    render(<WebviewSearch webviewRef={webviewRef} isWebviewReady appId="app-1" />)
-    await openSearchOverlay()
-
-    const input = screen.getByRole('textbox')
-    await user.type(input, 'Cherry')
-
-    await waitFor(() => {
-      expect(toastMock.error).toHaveBeenCalledWith('Error')
-    })
-  })
-
-  it('stops search when component unmounts', async () => {
-    const { stopFindInPageMock, webview } = createWebviewMock()
-    const webviewRef = { current: webview } as React.RefObject<WebviewTag | null>
-
-    const { unmount } = render(<WebviewSearch webviewRef={webviewRef} isWebviewReady appId="app-1" />)
-    await openSearchOverlay()
-
-    stopFindInPageMock.mockClear()
-    unmount()
-
-    expect(stopFindInPageMock).toHaveBeenCalledWith('clearSelection')
-  })
-
-  it('ignores keyboard shortcut when webview is not ready', async () => {
-    const { findInPageMock, webview } = createWebviewMock()
-    const webviewRef = { current: webview } as React.RefObject<WebviewTag | null>
-
-    render(<WebviewSearch webviewRef={webviewRef} isWebviewReady={false} appId="app-1" />)
-
-    await act(async () => {
-      fireEvent.keyDown(window, { key: 'f', ctrlKey: true })
-    })
-
-    expect(screen.queryByPlaceholderText('Search')).not.toBeInTheDocument()
-    expect(findInPageMock).not.toHaveBeenCalled()
-  })
-})
--- a/Show More
+++ b/Show More