DeepChat V0.3.8

🚀 DeepChat 0.3.8 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

主要变更 / Key Changes：

• 升级 TailwindCSS 到 v4，并移除了旧插件与配置，全面适配新版 Tailwind 语法和插件体系
• 移除了 Linux 版本的内置 node 和 uv 环境
• 依赖包全面升级，适配最新生态，包括 electron、@tailwindcss/typography、@tailwindcss/vite 等
• 统一组件引用样式，将自定义组件集成为 @shadcn 路径别名
• 优化 Tab 管理和拖拽新建逻辑，增强多窗口环境下的表现
• 移除部分冗余样式（如 body 背景，采用原生 CSS）
• uikit 新增 Think 内容块功能
• 修复了 Table 样式不正确的问题（#943），并对 markdown 相关依赖进行了升级
• 新增批量更新 shadcn 组件的脚本 update-shadcn.js
• 运行与构建脚本新增 VITE_ENABLE_PLAYGROUND 环境变量支持 playground 入口
• 优化多项 UI 细节，提高一致性与兼容性
• 修复知识库管理相关 bug，提升文件处理的健壮性
• 其它细节优化与体验提升

英文版 / English Version：

• Upgraded TailwindCSS to v4 with full migration to new syntax and plugins. Old plugin/config removed.
• Remove build-in node and uv in Linux version
• All dependencies updated to latest compatible versions, including electron, @tailwindcss/typography, and @tailwindcss/vite.
• Unified custom component imports via new @shadcn alias usage.
• Improved tab management and drag-to-create logic, with better experience in multi-window scenarios.
• Removed redundant styles (e.g., body background now uses native CSS).
• Added new Think content block feature in uikit.
• Fixed incorrect Table style (#943) and upgraded markdown-related dependencies.
• Added update-shadcn.js, a CLI script for batch updating shadcn components.
• Build and dev scripts now support the VITE_ENABLE_PLAYGROUND env for playground tab.
• Polished various UI details, enhancing consistency and cross-platform compatibility.
• Bugfixes for knowledge base file management to improve robustness.
• Miscellaneous refinements and UX improvements.

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

What's Changed

• fix: table style by @Simon-He95 in #943
• refactor: migration to tailwind v4 by @zerob13 in #940
• feat(uikit): add new think content block by @zerob13 in #947

Full Changelog: v0.3.7...v0.3.8

DeepChat V0.3.7

🚀 DeepChat 0.3.7 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

• Added support for Qwen3-VL and Qwen3-Max (0924), including new model configuration options and thinking budget logic.

• Upgraded vue-renderer-markdown to 0.0.54-beta.6, enhancing markdown rendering and custom component support.

• Removed DOMPurify dependency and refactored MessageBlockThink/HTMLArtifact security handling.

• Improved code organization: artifact processing logic, attribute configuration, and related refactors.

• Fixed various style issues for KaTeX and code blocks, optimizing UI consistency.

• Updated dependencies: added Katex, updated vue-use-monaco and other libraries for better compatibility.

• 新增 Qwen3-VL、Qwen3-Max (0924) 支持，包含模型配置项和思考预算相关逻辑。

• vue-renderer-markdown 升级至 0.0.54-beta.6，改进 Markdown 渲染和自定义组件支持。

• 移除 DOMPurify 依赖，重构 MessageBlockThink/HTMLArtifact 安全处理方案。

• 优化代码结构：重组 artifact 处理逻辑及属性配置相关重构。

• 修复 KaTeX 和代码块等样式问题，提升界面一致性体验。

• 依赖更新：新增 Katex，升级 vue-use-monaco 及其它相关库，增强兼容性。

（本版本为 Beta，欢迎试用并反馈问题。）

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

What's Changed

• fix: katex style by @zerob13 in #926
• refactor: reorganize artifact processing logic and improve attribute parsing by @Simon-He95 in #927
• fix(thinkingBlock): pre style by @Simon-He95 in #930
• refactor: remove DOMPurify dependency and replace MessageBlockThink v-html with vue-markdown-renderer by @Simon-He95 in #931
• feat: add Qwen3-VL and Qwen3-Max 0924 by @yyhhyyyyyy in #932
• refactor: update vue-renderer-markdown to 0.0.54-beta.6 and remove Re… by @Simon-He95 in #934
• Bugfix/custom prompt cant load content by @zerob13 in #936

Full Changelog: v0.3.6...v0.3.7

DeepChat V0.3.6

🚀 DeepChat 0.3.6 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

• Fix: Migrate model config data from new default settings

• Chore: Update Monaco editor dependencies

• Fix: Add support for zero temperature option

• Fix: Add 'copied' keys and remove copy toast

• Fix: Make model selector choose correct model when thread is pinned

• Feature: Add regenerate action on user messages

• Fix: Code block rendering bug in markdown renderer

• Fix: Layer support in vue-renderer-markdown

• Other optimizations, dependency bumps, and code cleanliness improvements

• 修复：模型配置数据迁移以适配新默认设置

• 其他：更新 Monaco 编辑器依赖

• 修复：新增零温度选项支持

• 修复：添加 copied 键并移除复制提示

• 修复：固定主题时模型选择器可选中正确模型

• 新增功能：用户消息支持重新生成

• 修复：Markdown 渲染器代码块渲染问题

• 修复：vue-renderer-markdown 支持图层

• 其他优化、依赖升级及代码清理完善

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

What's Changed

• fix: add missing qwen3-max search configuration in frontend by @yyhhyyyyyy in #877
• fix: upgrade reboot failed by @zerob13 in #876
• fix: auto-scroll to view after pasting large content by @yyhhyyyyyy in #878
• fix: preserve line breaks and formatting when pasting text in ChatInput by @yyhhyyyyyy in #879
• fix: align DashScope thinking models support between frontend and backend by @yyhhyyyyyy in #883
• refactor: replace mime-types with es-mime-types by @Simon-He95 in #884
• feat: add full-screen in mermaid & perf renderer code_block by @Simon-He95 in #885
• feat: add live search support for Grok provider by @yyhhyyyyyy in #886
• feat: add GoogleSearch support for Gemini provider by @neoragex2002 in #889
• feat: mcpclient support json $ref schema by @zerob13 in #890
• feat: add chatConfig entry for gemini search option by @neoragex2002 in #891
• fix: revert json ref support by @zerob13 in #892
• chore(deps-dev): bump vite from 7.1.4 to 7.1.5 by @dependabot[bot] in #893
• fix: add missing delete button for anthropic custom providers by @yyhhyyyyyy in #895
• chore: update vue-renderer-markdown & vue-use-monaco deps to perf use Monaco use raf render by @Simon-He95 in #896
• fix: paste image into editor not working by @zerob13 in #900
• fix: fork msg can not retry by @zerob13 in #902
• fix: add image capture for variants by @zerob13 in #904
• feat: upgrade vue-renderer-markdown support code_block collpase and mermaid progressive render by @Simon-He95 in #903
• chore: upgrade vue-renderer-markdown perf mermaid render by @Simon-He95 in #908
• fix: css style override by @zerob13 in #911
• fix: Enter inserts newline in edit mode by @yyhhyyyyyy in #912
• feat: add regenerate on user messages by @yyhhyyyyyy in #913
• fix: make model select choose right model when pinned thread by @zerob13 in #914
• fix: add 'copied' keys and drop copy toast by @yyhhyyyyyy in #917
• fix: add fix for setting page with shortcut by @zerob13 in #919
• feat: add support for zero temperature by @zerob13 in #920
• fix: migration model config data from new default settings by @zerob13 in #923

New Contributors

• @dependabot[bot] made their first contribution in #893

Full Changelog: v0.3.5...v0.3.6

DeepChat V0.3.5

🚀 DeepChat 0.3.5 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

⚠️ 重要通知：从本版本开始，团队将暂停接受 feature（功能类）PR，需等新的架构与 UI 发布后才会恢复。如有需求类 PR，请提前联系开发团队沟通。

Release Highlights / 主要变更

• 停接 feature 类 PR，专注架构与 UI 大重构
  后续仅处理修复与重构相关内容，feature 需求请与开发团队提前对接。
• 基础改进与修复
  • 聊天模型切换后自动同步最新配置
  • 适配与兼容多家模型供应商（如 OpenAI、Gemini Nano、Banana、Siliconcloud、DashScope、Qwen3、Grok 等），包括输入、推理支持与更多模型配置能力
  • 文件类型标记修正，避免 audio/HTML 误判
  • 拖拽窗口大小异常修复
  • MCP 工具和 prompt 管理优化，日志改进
  • 系统 prompt 增强（如插入当前日期时间）
  • UI/Markdown 滚动、按钮 glow、Clipboard 粘贴体验优化
• 代码和国际化提升
  • presenter/agent/type 强类型解耦与重构
  • 全面翻译中文注释与日志
  • 多语言包/i18n 内容补全及修正

---

⚠️ Notice: Starting from this release, feature PRs are temporarily suspended until the new architecture and UI are published. If you have any feature requests, please contact the dev team in advance.

Release Highlights

• Feature PRs on hold, focus on major refactor and UI overhaul
  Only bugfixes and refactor-related changes are accepted. Please reach out to the dev team early for any feature discussions.
• General improvements and fixes
  • Chat config now synchronizes when switching models
  • Enhanced adaptation and compatibility for multiple model providers (OpenAI, Gemini Nano, Banana, Siliconcloud, DashScope, Qwen3, Grok, etc.), including improved input, reasoning features, and more configurable options
  • File-type tag fix to avoid HTML/audio confusion
  • Fixed abnormal window resizing with drag
  • Optimized MCP prompt management and tool logging
  • Enhanced system prompt logic (e.g., current datetime insertion)
  • UI/Markdown scrolling, button glow, clipboard data sanitization improvements
• Code structure & i18n
  • Presenter/agent/type strong typing and modular refactor
  • All Chinese comments/logs translated to English
  • i18n package fixes and content enrichment

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

What's Changed

• fix: chat confg need sync to new value when change model in chat by @zerob13 in #823
• fix: remove file-type ,this will mark html as audio by @zerob13 in #824
• fix: Solve the problem of the window becoming larger when dragging floating button under Windows by @hllshiro in #826
• chore: update vue-renderer-markdown to v0.0.35 by @yyhhyyyyyy in #827
• refactor: remove custom-prompts-server by @zerob13 in #829
• feat: ux update by @zhangmo8 in #831
• feat: setting provider ux update by @zhangmo8 in #832
• fix(markdown): auto-wrap hide scroll by @Simon-He95 in #833
• feat: add enable_thinking parameter support for siliconcloud by @yyhhyyyyyy in #835
• chore: bump deps by @zerob13 in #834
• refactor: adjust scroll-to-bottom button glow effect by @yyhhyyyyyy in #837
• feat: add mutual exclusive confirmation dialogs for DeepSeek-V3.1 by @yyhhyyyyyy in #838
• feat: add sanitizeText utility for clipboard data handling by @Simon-He95 in #843
• feat: support canary upgrade by @zerob13 in #840
• feat: implement thinking parameter support for doubao models by @yyhhyyyyyy in #842
• feat: implement dedicated DashScope provider with enable_thinking support by @yyhhyyyyyy in #844
• fix: prioritize provider.id over apiType in createProviderInstance by @yyhhyyyyyy in #846
• feat: add qwen3 thinking budget support by @yyhhyyyyyy in #848
• feat: add qwen3 thinking budget support in ChatConfig by @yyhhyyyyyy in #849
• refactor(types): 🚀split monolithic presenter.d.ts into strict per-domain *.presenter.d.ts + typed core layer by @zerob13 in #847
• feat: add glow breathing effect to scroll-to-bottom button by @neoragex2002 in #850
• feat: add web search support with configurable options for dashscope by @yyhhyyyyyy in #851
• feat: add web search support with configurable options for dashscope by @yyhhyyyyyy in #852
• feat: add dashscope commercial models to enable_thinking support by @yyhhyyyyyy in #853
• feat: add search capability icon for model list by @yyhhyyyyyy in #854
• feat(markdown): Thinking panel now supports LaTeX compilation for mathematical formulas & markdown performance optimization by @Simon-He95 in #857
• feat: upgrade vue-renderer-markdown & vue-use-monaco by @Simon-He95 in #862
• feat(dashscope): add qwen3-max-preview model by @yyhhyyyyyy in #865
• fix: mcp params support more types by @zerob13 in #861
• fix: prevent @ symbol remaining when deleting mentions by @yyhhyyyyyy in #867
• feat: implement separated system and custom prompt management by @yyhhyyyyyy in #868
• feat: add Moonshot model configurations by @yyhhyyyyyy in #869
• refactor: translate all cn comments and log to en by @zerob13 in #871
• feat: add reasoning support for Grok thinking models by @yyhhyyyyyy in #873

Full Changelog: v0.3.4...v0.3.5

DeepChat V0.3.4

🚀 DeepChat 0.3.4 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

新功能

• 支持 TokenFlux.ai、更新豆包模型
• 全新启动画面 + 主题
• 悬浮按钮 JS 重写，支持多屏边界检测

优化

• Gemini 推理解析、模型配置、关闭内存占用
• 减少 IPC 调用、并行退出、原子化 provider 操作
• Markdown 转义、导入器修复
• 优化了默认的系统提示词，并且允许用户在设置中进行修改

New

• TokenFlux.ai provider + updated Doubao models
• New splash screen with themes
• Floating button re-written in JS, multi-monitor aware

Improved

• Gemini reasoning & model fixes, memory leak on close
• Fewer IPC calls, faster shutdown, atomic provider ops
• Markdown escape fixes, importer tweaks
• Optimized the default system prompts and allowed users to modify the built-in system prompts

What's Changed

• fix: Correctly parse Gemini reasoning for both new thought field an… by @neoragex2002 in #801
• fix: correct Gemini 2.5 Flash Image Preview model capabilities and token limits by @yyhhyyyyyy in #802
• feat: add TokenFlux.ai provider support by @zerob13 in #804
• fix: optimize polling logic to prevent unnecessary IPC calls by @yyhhyyyyyy in #806
• fix: escape brackets in Markdown content to prevent parsing issues & refactor MessageBlockContent if branch by @Simon-He95 in #808
• fix memory when close window by @hllshiro in #809
• Reorganize gemini model configs; Refine Gemini reasoning content parsing; Markdown rendering of fixed by @neoragex2002 in #807
• feat: refactor floating button with JavaScript-based drag and fix interaction issues by @yyhhyyyyyy in #810
• Refactor/app lifecycle management by @hllshiro in #803
• perf: parallelize lifecycle hooks execution for faster app shutdown by @yyhhyyyyyy in #811
• fix: resolve race condition in floating button drag-and-drop detection by @neoragex2002 in #813
• fix: resolve floating button multi-display boundary detection issue by @yyhhyyyyyy in #814
• Optimize/lifecycle actuator optimization by @hllshiro in #812
• perf: implement atomic provider operations to optimize performance by @yyhhyyyyyy in #815
• feat: update Doubao provider with new models by @yyhhyyyyyy in #816
• fix: resolve application hang when quitting during startup by @yyhhyyyyyy in #817
• chore: fix DataImporter bug and some ux optimises by @zerob13 in #818

Full Changelog: v0.3.3...v0.3.4

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

DeepChat V0.3.3

🚀 DeepChat 0.3.3

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

Highlights

• Added support for OpenRouter Gemini 2.5 flash image preview
• Enhanced image file display with context-aware thumbnails
• Improved scroll-to-bottom button loading animation with a glowing effect
• Optimized image markdown streaming performance using a smart buffering mechanism
• Increased content length limit for web retrieval to capture more data

Fixes

• Fixed CSS bundle issue for better styling consistency
• Corrected provider icon alignment in message list
• Synced message navigation sidebar with latest variant content
• Improved auto-scroll behavior for fast AI responses
• Resolved session state conflict between floating and main windows

Maintenance

• Version bumped to 0.3.3

---

发布说明 (2025年8月26日 - 8月28日)

亮点

• 支持OpenRouter Gemini 2.5闪存镜像预览
• 增强图片文件显示，支持上下文感知缩略图
• 优化“滚动到底部”按钮加载动画，增加发光效果
• 优化图片Markdown流式加载性能，采用智能缓冲机制
• 提高网页内容抓取的长度限制，捕获更多数据

修复

• 修复CSS打包问题，提升样式一致性
• 修正消息列表中提供者图标对齐问题
• 同步消息导航侧边栏与最新变体内容
• 改善快速AI响应时的自动滚动行为
• 解决浮动窗口与主窗口间的会话状态冲突

维护

• 版本更新至0.3.3

What's Changed

• fix: correct provider icon alignment in message list by @yyhhyyyyyy in #789
• feat: support openrouter gemini 2.5 flash image preview by @yyhhyyyyyy in #791
• feat: enhance image file display with context-aware thumbnails by @yyhhyyyyyy in #793
• feat: improve scroll-to-bottom button loading animation with glow effect by @yyhhyyyyyy in #794
• fix: sync message navigation sidebar with latest variant content by @yyhhyyyyyy in #796
• feat: optimize image markdown streaming performance with smart buffering by @zerob13 in #792
• feat: Increase the content length limit for web retrieval to capture more context and provide more accurate responses by @Dw9 in #797
• fix: improve auto scroll behavior for fast AI responses by @yyhhyyyyyy in #798
• fix: resolve session state conflict between floating and main windows by @yyhhyyyyyy in #799

Full Changelog: v0.3.2...v0.3.3

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

DeepChat V0.3.2

🚀 DeepChat 0.3.2 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

✨ 新功能 / New Features

1. AWS Bedrock 模型接入
   原生支持调用 Amazon Bedrock 的全部模型家族，包含新的设置面板与区域/凭据管理。
   Native support for the entire Amazon Bedrock model family, with a dedicated settings panel for regions & credentials.

2. HTML 预览 – 多尺寸视图
   渲染 HTML 内容时，可一键切换桌面 / 平板 / 手机三种尺寸，实时预览响应式效果。
   Responsive HTML preview now supports desktop, tablet and mobile viewports.

3. 消息侧边栏导航与全文搜索
   左侧新增可折叠的消息导航面板；支持关键字高亮、即时过滤，以及快速跳转到指定消息。
   New collapsible message navigation sidebar with full-text search, highlighting & quick-jump.

4. 自动滚动与“回到最新”按钮
   生成消息时自动保持最新内容可见；一键滚动到底部，并带有加载动画。
   Auto-scroll to latest message during generation + animated “scroll-to-bottom” button.

5. 品牌焕新 (Rebrand) 支持
   全量替换应用标识、名称、主题色等，并配套完整的中英文迁移文档。
   Complete rebrand kit: swap app identity, name, theme colors + bilingual migration guide.

---

🔧 改进 / Enhancements

• 内置知识切片算法重构
  更快的分块速度、更低的内存占用、对中文/代码混合文本更友好。
  Refactored built-in knowledge text splitter for faster & memory-efficient chunking, better CJK/code mixed text support.

• 依赖升级
  better-sqlite3-multiple-ciphers → v12.2.0，已适配 Electron 37。
  Dependency bump for Electron 37 compatibility.

• 所有搜索引擎均可在设置内一键测试连通性
  解决因验证码/网络差异导致的配置困扰。
  Every search engine can now be connectivity-tested inside settings to preempt CAPTCHA issues.

---

🐞 问题修复 / Bug Fixes

• UI/样式

  • 修复 CSS 样式注入顺序导致的优先级错乱
  • 修复 MCP 市场按钮对齐及布局异常 (#769)
  • 修复主题切换后部分组件未即时更新 (#770)
  • 消息导航滚动定位现在始终可见 (#776)
  • 字体大小变更实时同步到所有标签页 (#786)
  • 新增图标支持，避免缺失导致的异常空白

• 模型与 Provider

  • 禁用 deepseek-reasoner 的原生工具调用，恢复思考链展示
  • 修复 Gemini 2.5 Thinking Budget 初始化与跨标签同步问题
  • 修复模型列表、Provider 顺序在多个标签页间的同步异常 (#763, #765)

• 交互体验

  • 生成结束后不再强制滚动到底部 (#783)
  • 正则表达式移除多余转义字符，避免匹配错误 (#764)
  • 修复 Windows 下 VS Redistributable 检测提示重复出现 (#755)

---

What's Changed

• fix vsredist check nsh by @hllshiro in #755
• Optimize/builtin knowledge text splitter by @hllshiro in #757
• feat: add aws bedrock provider by @bigyao25 in #756
• fix: improve AWS Bedrock provider settings UI by @yyhhyyyyyy in #759
• feat: html预览新增尺寸切换 by @Yealove in #618
• chore: bump better-sqlite3-multiple-ciphers to 12.2.0 for Electron 37 by @zerob13 in #761
• fix: resolve model list synchronization issue across tabs by @yyhhyyyyyy in #763
• fix: remove unnecessary escape character in regex pattern by @yyhhyyyyyy in #764
• fix: sync provider order changes across all tabs by @yyhhyyyyyy in #765
• fix: resolve Gemini 2.5 thinking budget initialization and sync issues by @yyhhyyyyyy in #767
• fix: resolve MCP market button alignment and layout issues by @yyhhyyyyyy in #769
• fix: theme change issue by @zerob13 in #770
• fix: enable search engine testing for all engines to resolve CAPTCHA timeout issue by @zerob13 in #771
• feat: add message navigation sidebar with search functionality by @yyhhyyyyyy in #776
• fix: improve message navigation scroll position to show content from top by @yyhhyyyyyy in #777
• feat: support rebrand by @zerob13 in #778
• feat: implement auto-scroll to latest message and scroll-to-bottom button by @yyhhyyyyyy in #780
• doc: optimize and remove outdated docs by @zerob13 in #781
• fix: prevent automatic scroll to bottom after message generation by @yyhhyyyyyy in #783
• feat: add loading animation to scroll-to-bottom button during generation by @yyhhyyyyyy in #784
• fix: sync font size changes across all tabs in real-time by @yyhhyyyyyy in #786
• feat: replace font size slider with button group for better UX by @yyhhyyyyyy in #788

New Contributors

• @Yealove made their first contribution in #618

Full Changelog: v0.3.1...v0.3.2

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

DeepChat V0.3.1

🚀 DeepChat 0.3.1 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

• 修复了内存泄漏问题

• 修复了DeepLink 和 Artifacts 带来的安全漏洞

• 为提供商设置添加搜索功能

• 为 API 密钥输入添加密码可见性切换

• 添加了Vercel AI 和AWS Bedrock 新供应商

• 为 OpenRouter 添加全面的 OpenAI 推理模型支持

• 增加了GPT5系列模型的更完善支持

• 为 GPT-OSS 模型添加 reasoning_effort 参数支持

• 添加架构兼容性检查和 Visual C++ Redistribu 支持

• 解决使用 Ctrl+Shift+N 创建新窗口时的焦点闪烁问题

• 为 PopoverContent 添加可滚动支持以防止溢出

• Fixed memory leak issues

• Fixed security vulnerabilities caused by DeepLink and Artifacts

• Added search functionality for provider settings

• Added password visibility toggle for API key inputs

• Added new providers: Vercel AI and AWS Bedrock

• Added comprehensive OpenAI reasoning models support for OpenRouter

• Enhanced support for GPT-5 series models

• Added reasoning_effort parameter support for GPT-OSS models

• Added architecture compatibility checks and Visual C++ Redistributable support

• Resolved focus flicker issue when creating new windows with Ctrl+Shift+N

• Added scrollable support to PopoverContent to prevent overflow

What's Changed

• fix: resolve OpenRouter model sync timeout and welcome view UI issues by @yyhhyyyyyy in #738
• feat: add GLM-4.5V model support by @yyhhyyyyyy in #739
• feat: add search functionality for provider settings by @yyhhyyyyyy in #740
• fix: optimize ModelScope MCP sync UI layout for consistent window height by @yyhhyyyyyy in #741
• fix: resolve reasoningEffort parameter transmission issues by @yyhhyyyyyy in #742
• feat: add Vercel AI Gateway provider support by @yyhhyyyyyy in #743
• fix: pass user reasoningEffort/verbosity settings to GPT-5 API requests by @yyhhyyyyyy in #746
• feat: add password visibility toggle for API key inputs by @yyhhyyyyyy in #747
• Feature/add aws bedrock provider by @bigyao25 in #744
• fix: extend reasoningEffort support to all reasoning models by @yyhhyyyyyy in #749
• feat: add comprehensive OpenAI reasoning models support for OpenRouter by @yyhhyyyyyy in #751
• chore(deps): bump electron to 37.3.0 fix #737 by @zerob13 in #753
• feat: add architecture compatibility checks and Visual C++ Redistributable installation support by @hllshiro in #752

Full Changelog: v0.3.0...v0.3.1

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

DeepChat V0.3.0

🚀 DeepChat 0.3.0 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

• ✅ 新增模型：GPT-5 系列、Claude Opus 4.1

• ✅ 悬浮对话：浮动窗口、ESC 关闭、位置记忆

• ✅ MCP 生态：市场同步、魔搭社区接入

• ⚡ 性能提升：单文件打包、窗口创建更流畅

• 🔧 体验优化：焦点闪烁、数学渲染、文件移动问题修复

• 📚 文档更新：开发者链接、翻译优化

• ✅ New models: GPT-5 series, Claude Opus 4.1

• ✅ Hovering Dialogue: Floating window, ESC off, position memory

• ✅ MCP Ecosystem: Market synchronization and community access

• ⚡ Performance improvements: Single file packaging and smoother window creation

• 🔧 Experience Optimization: Fix focus flickering, mathematical rendering, file movement issues

• 📚 Documentation updates: developer links, translation optimizations

What's Changed

• fix: add AlertDialogDescription to resolve accessibility warning by @yyhhyyyyyy in #706
• fix: resolve focus flicker when creating new windows with Ctrl+Shift+N by @yyhhyyyyyy in #707
• fix: memory leak issue when closing window by @hllshiro in #709
• docs: update zhipu developer doc website link by @tomsun28 in #715
• refactor: better translate by @zerob13 in #716
• feat: add GPT-5 series model support by @yyhhyyyyyy in #717
• ci(vite): Bundle the main file into a single file to speed up loading. by @zerob13 in #718
• fix(math): parser by upgrade vue-renderer-markdown by @Simon-He95 in #722
• chore: bump deps by @zerob13 in #721
• fix: add scrollable support to PopoverContent to prevent overflow by @yyhhyyyyyy in #720
• feat: implement floating chat window system with performance optimization by @yyhhyyyyyy in #724
• feat: add mcp sync and modelscope provider #615 by @zerob13 in #723
• fix: move_files newPath 参数计算规则 by @wangle201210 in #725
• feat: add Claude Opus 4.1 to anthropic default model list by @yyhhyyyyyy in #726
• feat: Add mcprouter's MCP marketplace api support by @zerob13 in #727
• feat: add ESC key to close floating chat window by @yyhhyyyyyy in #728
• feat: add floating button position persistence with boundary validation by @yyhhyyyyyy in #729
• feat: add reasoning_effort parameter support for gpt-oss models by @yyhhyyyyyy in #731
• fix: artifacts code not streaming by @zerob13 in #732
• feat: disable automatic model enabling for better UX by @yyhhyyyyyy in #734
• feat: sync provider sorting from settings to model selection by @yyhhyyyyyy in #736

New Contributors

• @tomsun28 made their first contribution in #715

Full Changelog: v0.2.9...v0.3.0
🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！

DeepChat V0.2.9

🚀 DeepChat 0.2.9 正式发布 | 重新定义你的 AI 对话体验！

—— 不再是简单的 ChatBot，而是你的自然语言 Agent 工具🌟

主要更新

• 基于服务商的请求速率限制： 告别 too many request，自定义各个供应商的请求限制
• 适配了 Claude Opus 4.1和 OpenAI gpt-oss
• 性能优化，启动速度变快了一点点
• 常规 Bug 修复

What's Changed

• fix: prevent tooltip from overlapping HTML preview in ArtifactDialog by @yyhhyyyyyy in #683
• fix: unwatch is not a function error in knowledge settings components by @yyhhyyyyyy in #684
• fix: resolve message editing and markdown rendering issues by @yyhhyyyyyy in #685
• docs: update Electron-Vite link format to full URL by @cycleccc in #687
• fix: built-in knowledge settings not loaded when first entering by @hllshiro in #688
• fix: resolve database lock issue in data reset functionality by @yyhhyyyyyy in #689
• feat: implement provider-based request rate limit by @yyhhyyyyyy in #686
• Optimize/builtin knowledge interaction by @hllshiro in #690
• fix: fixed dirty data problems by @hllshiro in #691
• feat: add mentions support in deeplink by @hllshiro in #692
• bugfix: prompt attach files by @hllshiro in #695
• feat: add npm registry caching and optimization system by @yyhhyyyyyy in #697
• fix: resolve input focus loss issue after sending messages by @yyhhyyyyyy in #694
• fix: resolve deeplink message not displaying when triggered from settings page by @yyhhyyyyyy in #699
• fix: a potential problem with mcp tools by @hllshiro in #698
• feat: add support for new models by @zerob13 in #700
• fix: refine #698 by @hllshiro in #701
• feat: add Claude Opus 4.1 support and fix anthropic model list dialog by @yyhhyyyyyy in #702

New Contributors

• @cycleccc made their first contribution in #687

Full Changelog: v0.2.8...v0.2.9

🔥 为什么选择 DeepChat？

✅ 商业友好：基于原版 Apache License 2.0 开源，无任何协议外的额外约束，面向开源。
✅ 开箱即用：极简配置，即刻开启你的智能对话之旅。
✅ 极致灵活：自由切换模型，自定义模型源，满足你多样化的对话和探索需求。
✅ 体验绝佳：LaTeX 公式渲染、代码高亮、Markdown 支持，模型对话从未如此顺畅。
✅ 持续进化：我们倾听用户反馈，不断迭代更新，为你带来更卓越的 AI 对话体验。

📥 立即体验未来

💬 反馈有礼：欢迎提交你的宝贵建议，加入 VIP 用户社群，与我们一同塑造 DeepChat 的未来！