🚀 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