mirror of
https://github.com/CherryHQ/cherry-studio.git
synced 2025-12-19 06:30:10 +08:00
chore: release v1.7.4
🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
parent
d41229c69b
commit
b66787280a
@ -134,60 +134,54 @@ artifactBuildCompleted: scripts/artifact-build-completed.js
|
||||
releaseInfo:
|
||||
releaseNotes: |
|
||||
<!--LANG:en-->
|
||||
Cherry Studio 1.7.3 - Feature & Stability Update
|
||||
Cherry Studio 1.7.4 - New Browser MCP & Model Updates
|
||||
|
||||
This release brings new features, UI improvements, and important bug fixes.
|
||||
This release adds a powerful browser automation MCP server, new web search provider, and model support updates.
|
||||
|
||||
✨ New Features
|
||||
- Add MCP server log viewer for better debugging
|
||||
- Support custom Git Bash path configuration
|
||||
- Add print to PDF and save as HTML for mini program webviews
|
||||
- Add CherryIN API host selection settings
|
||||
- Enhance assistant presets with sort and batch delete modes
|
||||
- Open URL directly for SelectionAssistant search action
|
||||
- Enhance web search tool switching with provider-specific context
|
||||
- [MCP] Add @cherry/browser CDP MCP server with session management for browser automation
|
||||
- [Web Search] Add ExaMCP free web search provider (no API key required)
|
||||
- [Model] Support GPT 5.2 series models
|
||||
- [Model] Add capabilities support for Doubao Seed Code models (tool calling, reasoning, vision)
|
||||
|
||||
🔧 Improvements
|
||||
- Remove Intel Ultra limit for OVMS
|
||||
- Improve settings tab and assistant item UI
|
||||
- [Translate] Add reasoning effort option to translate service
|
||||
- [i18n] Improve zh-TW Traditional Chinese locale
|
||||
- [Settings] Update MCP Settings layout and styling
|
||||
|
||||
🐛 Bug Fixes
|
||||
- Fix stack overflow with base64 images
|
||||
- Fix infinite loop in knowledge queue processing
|
||||
- Fix quick panel closing in multiple selection mode
|
||||
- Fix thinking timer not stopping when reply is aborted
|
||||
- Fix ThinkingButton icon display for fixed reasoning mode
|
||||
- Fix knowledge query prioritization and intent prompt
|
||||
- Fix OpenRouter embeddings support
|
||||
- Fix SelectionAction window resize on Windows
|
||||
- Add gpustack provider support for qwen3 thinking mode
|
||||
- [Chat] Fix line numbers being wrongly copied from code blocks
|
||||
- [Translate] Fix default to first supported reasoning effort when translating
|
||||
- [Chat] Fix preserve thinking block in assistant messages
|
||||
- [Web Search] Fix max search result limit
|
||||
- [Embedding] Fix embedding dimensions retrieval for ModernAiProvider
|
||||
- [Chat] Fix token calculation in prompt tool use plugin
|
||||
- [Model] Fix Ollama provider options for Qwen model support
|
||||
- [UI] Fix Chat component marginRight calculation for improved layout
|
||||
|
||||
<!--LANG:zh-CN-->
|
||||
Cherry Studio 1.7.3 - 功能与稳定性更新
|
||||
Cherry Studio 1.7.4 - 新增浏览器 MCP 与模型更新
|
||||
|
||||
本次更新带来新功能、界面改进和重要的问题修复。
|
||||
本次更新新增强大的浏览器自动化 MCP 服务器、新的网页搜索提供商以及模型支持更新。
|
||||
|
||||
✨ 新功能
|
||||
- 新增 MCP 服务器日志查看器,便于调试
|
||||
- 支持自定义 Git Bash 路径配置
|
||||
- 小程序 webview 支持打印 PDF 和保存为 HTML
|
||||
- 新增 CherryIN API 主机选择设置
|
||||
- 助手预设增强:支持排序和批量删除模式
|
||||
- 划词助手搜索操作直接打开 URL
|
||||
- 增强网页搜索工具切换逻辑,支持服务商特定上下文
|
||||
- [MCP] 新增 @cherry/browser CDP MCP 服务器,支持会话管理的浏览器自动化
|
||||
- [网页搜索] 新增 ExaMCP 免费网页搜索提供商(无需 API 密钥)
|
||||
- [模型] 支持 GPT 5.2 系列模型
|
||||
- [模型] 为豆包 Seed Code 模型添加能力支持(工具调用、推理、视觉)
|
||||
|
||||
🔧 功能改进
|
||||
- 移除 OVMS 的 Intel Ultra 限制
|
||||
- 优化设置标签页和助手项目 UI
|
||||
- [翻译] 为翻译服务添加推理强度选项
|
||||
- [国际化] 改进繁体中文(zh-TW)本地化
|
||||
- [设置] 优化 MCP 设置布局和样式
|
||||
|
||||
🐛 问题修复
|
||||
- 修复 base64 图片导致的栈溢出问题
|
||||
- 修复知识库队列处理的无限循环问题
|
||||
- 修复多选模式下快捷面板意外关闭的问题
|
||||
- 修复回复中止时思考计时器未停止的问题
|
||||
- 修复固定推理模式下思考按钮图标显示问题
|
||||
- 修复知识库查询优先级和意图提示
|
||||
- 修复 OpenRouter 嵌入模型支持
|
||||
- 修复 Windows 上划词助手窗口大小调整问题
|
||||
- 为 gpustack 服务商添加 qwen3 思考模式支持
|
||||
- [聊天] 修复代码块中行号被错误复制的问题
|
||||
- [翻译] 修复翻译时默认使用第一个支持的推理强度
|
||||
- [聊天] 修复助手消息中思考块的保留问题
|
||||
- [网页搜索] 修复最大搜索结果数限制
|
||||
- [嵌入] 修复 ModernAiProvider 嵌入维度获取问题
|
||||
- [聊天] 修复提示词工具使用插件的 token 计算问题
|
||||
- [模型] 修复 Ollama 提供商对 Qwen 模型的支持选项
|
||||
- [界面] 修复聊天组件右边距计算以改善布局
|
||||
<!--LANG:END-->
|
||||
|
||||
@ -1,6 +1,6 @@
|
||||
{
|
||||
"name": "CherryStudio",
|
||||
"version": "1.7.3",
|
||||
"version": "1.7.4",
|
||||
"private": true,
|
||||
"description": "A powerful AI assistant for producer.",
|
||||
"main": "./out/main/index.js",
|
||||
|
||||
Loading…
Reference in New Issue
Block a user