介绍
Cherry Studio 是一款跨平台桌面客户端，支持多个大语言模型（LLM）服务商，兼容 Windows、Mac 和 Linux 系统，并拥丰富的个性化选项与领先的功能设计。
软件截图
软件特点
😌 轻松上手
🍏Windows，Mac，Linux跨平台支持
📦开箱即用，无需 Python 与 Docker
🤝简洁、友好的界面与交互设计
🛠️多样化的 LLM 服务模式支持
☁️ 全面覆盖 LLM 云服务，支持自定义 api key 与模型管理：OpenAI，Gemini，Anthropic，硅基流动...
🔗汇聚流行的 AI Web 服务，并计划通过功能增强提升体验：Claude，Peplexity，Poe，腾讯元宝，知乎直答...
💻支持 Ollama 运行本地模型
📲个性化的功能体验
📄完整的 Markdown 与 Mermaid 渲染支持
🤖使用与创建智能体提升工作效率
🔤持续迭代的翻译功能
🤲生成结果支持 Markdown 与图片分享
📎文件与图片上传，RAG 与多模态对话
🎨透明窗口与明暗主题支持
更新日志
v1.7.3
fix(windows): add manual window resize for SelectionAction window by @0xfullex in #11766
fix: add support for OpenRouter embeddings in listModels method by @DeJeune in #11774
fix(models): include model name as fallback for id field by @EurFelux in #11760
fix(aiCore): correct provider adaptation with model parameter by @EurFelux in #11758
fix: restore API version control with trailing # delimiter (addresses #11750) by @EurFelux in #11773
⬆️ chore(deps): upgrade @anthropic-ai/claude-agent-sdk to 0.1.62 by @vaayne in #11824
fix(aiCore): omit empty content in assistant messages with tool_calls by @zane0307-pixel in #11818
✨ feat: add MCP server log viewer by @vaayne in #11826
feat: add CherryIN API host selection settings by @kangfenmao in #11797
fix(knowledge): prioritize query & refine intent prompt by @Ying-xi in #11828
fix(ThinkingButton): show correct icon when isFixedReasoning by @EurFelux in #11825
chore: add gitcode release sync workflow by @defi-failure in #11807
feat(SelectionAssistant): open URL for search action by @0xfullex in #11770
feat: enhance web search tool switching logic to support provider-specific context by @DeJeune in #11769
🐛 fix: stabilize MCP log IPC registration by @vaayne in #11830
feat: Support custom git bash path by @beyondkmp in #11813
refactor(ui): improve settings tab and assistant item UI by @kangfenmao in #11819
fix: stop thinking timer when reply is aborted by @CalWade in #11794
feat: Add print to PDF and save as HTML for mini program webviews by @Copilot in #11104
feat(assistants): enhance ManageAssistantPresetsPopup with sort and batch delete modes by @kangfenmao in #11835
fix: add gpustack provider for qwen3 enable think by @DeJeune in #11843
feat: OVMS remove intel ultra limit by @makejiang in #11854
fix: infinite loop in knowledge queue processing by @defi-failure in #11856
fix: remove cloneDeep to prevent stack overflow with base64 images by @Copilot in #11761