[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-Mintplex-Labs--anything-llm":3,"tool-Mintplex-Labs--anything-llm":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",157379,2,"2026-04-15T23:32:42",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",108322,"2026-04-10T11:39:34",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":32,"last_commit_at":50,"category_tags":51,"status":17},6121,"gemini-cli","google-gemini\u002Fgemini-cli","gemini-cli 是一款由谷歌推出的开源 AI 命令行工具，它将强大的 Gemini 大模型能力直接集成到用户的终端环境中。对于习惯在命令行工作的开发者而言，它提供了一条从输入提示词到获取模型响应的最短路径，无需切换窗口即可享受智能辅助。\n\n这款工具主要解决了开发过程中频繁上下文切换的痛点，让用户能在熟悉的终端界面内直接完成代码理解、生成、调试以及自动化运维任务。无论是查询大型代码库、根据草图生成应用，还是执行复杂的 Git 操作，gemini-cli 都能通过自然语言指令高效处理。\n\n它特别适合广大软件工程师、DevOps 人员及技术研究人员使用。其核心亮点包括支持高达 100 万 token 的超长上下文窗口，具备出色的逻辑推理能力；内置 Google 搜索、文件操作及 Shell 命令执行等实用工具；更独特的是，它支持 MCP（模型上下文协议），允许用户灵活扩展自定义集成，连接如图像生成等外部能力。此外，个人谷歌账号即可享受免费的额度支持，且项目基于 Apache 2.0 协议完全开源，是提升终端工作效率的理想助手。",100752,"2026-04-10T01:20:03",[52,13,15,14],"插件",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":32,"last_commit_at":59,"category_tags":60,"status":17},4721,"markitdown","microsoft\u002Fmarkitdown","MarkItDown 是一款由微软 AutoGen 团队打造的轻量级 Python 工具，专为将各类文件高效转换为 Markdown 格式而设计。它支持 PDF、Word、Excel、PPT、图片（含 OCR）、音频（含语音转录）、HTML 乃至 YouTube 链接等多种格式的解析，能够精准提取文档中的标题、列表、表格和链接等关键结构信息。\n\n在人工智能应用日益普及的今天，大语言模型（LLM）虽擅长处理文本，却难以直接读取复杂的二进制办公文档。MarkItDown 恰好解决了这一痛点，它将非结构化或半结构化的文件转化为模型“原生理解”且 Token 效率极高的 Markdown 格式，成为连接本地文件与 AI 分析 pipeline 的理想桥梁。此外，它还提供了 MCP（模型上下文协议）服务器，可无缝集成到 Claude Desktop 等 LLM 应用中。\n\n这款工具特别适合开发者、数据科学家及 AI 研究人员使用，尤其是那些需要构建文档检索增强生成（RAG）系统、进行批量文本分析或希望让 AI 助手直接“阅读”本地文件的用户。虽然生成的内容也具备一定可读性，但其核心优势在于为机器",93400,"2026-04-06T19:52:38",[52,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":76,"owner_email":77,"owner_twitter":78,"owner_website":79,"owner_url":80,"languages":81,"stars":107,"forks":108,"last_commit_at":109,"license":110,"difficulty_score":32,"env_os":111,"env_gpu":112,"env_ram":113,"env_deps":114,"category_tags":121,"github_topics":123,"view_count":32,"oss_zip_url":76,"oss_zip_packed_at":76,"status":17,"created_at":143,"updated_at":144,"faqs":145,"releases":174},8003,"Mintplex-Labs\u002Fanything-llm","anything-llm","The all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.","AnythingLLM 是一款主打隐私与便捷的全能型 AI 生产力工具，旨在让用户在本地轻松构建专属的私有化大语言模型应用。它完美解决了用户在使用公共 AI 服务时面临的数据隐私担忧、复杂的部署配置以及高昂的使用门槛等痛点。\n\n通过 AnythingLLM，用户可以直观地“与文档对话”，将各类文件转化为可交互的知识库，还能利用 AI 智能体自动化处理复杂工作流。其独特的技术亮点在于“开箱即用”的设计理念：默认在本地设备运行，无需繁琐的环境搭建或参数调优，同时支持多用户协作与高度定制化配置，既保障了数据安全，又提供了企业级的功能深度。\n\n这款工具非常适合注重数据隐私的企业团队、希望私有化部署 AI 知识的科研人员，以及不愿被复杂技术劝退的普通用户。无论是开发者想要快速验证本地模型，还是职场人士希望安全地分析内部文档，AnythingLLM 都能提供一个无需妥协、功能完备且运行流畅的一站式解决方案，让每个人都能拥有自己的私人 AI 助手。","\u003Ca name=\"readme-top\">\u003C\u002Fa>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fanythingllm.com\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_e38b611303bd.png\" alt=\"AnythingLLM logo\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cdiv align='center'>\n\u003Ca href=\"https:\u002F\u002Ftrendshift.io\u002Frepositories\u002F2415\" target=\"_blank\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_4a68feb902da.png\" alt=\"Mintplex-Labs%2Fanything-llm | Trendshift\" style=\"width: 250px; height: 55px;\" width=\"250\" height=\"55\"\u002F>\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n\u003Cp align=\"center\">\n    \u003Cb>AnythingLLM:\u003C\u002Fb> The all-in-one AI app you were looking for.\u003Cbr \u002F>\n    Chat with your docs, use AI Agents, hyper-configurable, multi-user, & no frustrating setup required.\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002F6UyHPeGZAC\" target=\"_blank\">\n      \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fchat-mintplex_labs-blue.svg?style=flat&logo=data:image\u002Fpng;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAAAIGNIUk0AAHomAACAhAAA+gAAAIDoAAB1MAAA6mAAADqYAAAXcJy6UTwAAAH1UExURQAAAP\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002Fr6+ubn5+7u7\u002F3+\u002Fv39\u002Fenq6urq6\u002Fv7+97f39rb26eoqT1BQ0pOT4+Rkuzs7cnKykZKS0NHSHl8fdzd3ejo6UxPUUBDRdzc3RwgIh8jJSAkJm5xcvHx8aanqB4iJFBTVezt7V5hYlJVVuLj43p9fiImKCMnKZKUlaaoqSElJ21wcfT09O3u7uvr6zE0Nr6\u002FwCUpK5qcnf7+\u002Fnh7fEdKTHx+f0tPUOTl5aipqiouMGtubz5CRDQ4OsTGxufn515hY7a3uH1\u002FgXBydIOFhlVYWvX29qaoqCQoKs7Pz\u002FPz87\u002FAwUtOUNfY2dHR0mhrbOvr7E5RUy8zNXR2d\u002Ff39+Xl5UZJSx0hIzQ3Odra2\u002Fz8\u002FGlsbaGjpERHSezs7L\u002FBwScrLTQ4Odna2zM3Obm7u3x\u002FgKSmp9jZ2T1AQu\u002Fv71pdXkVISr2+vygsLiInKTg7PaOlpisvMcXGxzk8PldaXPLy8u7u7rm6u7S1tsDBwvj4+MPExbe4ueXm5s\u002FQ0Kyf7ewAAAAodFJOUwAABClsrNjx\u002FQM2l9\u002F7lhmI6jTB\u002FkA1GgKJN+nea6vy\u002FMLZQYeVKK3rVA5tAAAAAWJLR0QB\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\u002FBGf3cqCCiZOcnCe3QQIKHNRTpk6bDgpZjRkzg3pBQTBrdtCcuZCgluAD0vPmL1gIdvSixUuWgqNs2YJ+DUhkEYxuggkGmOQUcckrioPTJCOXEnZ5JS5YslbGnuyVERlDDFvGEUPOWvwqaH6RVkHKeuDMK6SKnHlVhTgx8jeTmqy6Eij7K6nLqiGyPwChsa1MUrnq1wAAACV0RVh0ZGF0ZTpjcmVhdGUAMjAyMy0xMC0wNFQwMDozODo0OSswMDowMB9V0a8AAAAldEVYdGRhdGU6bW9kaWZ5ADIwMjMtMTAtMDRUMDA6Mzg6NDkrMDA6MDBuCGkTAAAAKHRFWHRkYXRlOnRpbWVzdGFtcAAyMDIzLTEwLTA0VDAwOjM4OjQ5KzAwOjAwOR1IzAAAAABJRU5ErkJggg==\" alt=\"Discord\">\n  \u003C\u002Fa> |\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fblob\u002Fmaster\u002FLICENSE\" target=\"_blank\">\n      \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fstatic\u002Fv1?label=license&message=MIT&color=white\" alt=\"License\">\n  \u003C\u002Fa> |\n  \u003Ca href=\"https:\u002F\u002Fdocs.anythingllm.com\" target=\"_blank\">\n    Docs\n  \u003C\u002Fa> |\n   \u003Ca href=\"https:\u002F\u002Fmy.mintplexlabs.com\u002Faio-checkout?product=anythingllm\" target=\"_blank\">\n    Hosted Instance\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cb>English\u003C\u002Fb> · \u003Ca href='.\u002Flocales\u002FREADME.zh-CN.md'>简体中文\u003C\u002Fa> · \u003Ca href='.\u002Flocales\u002FREADME.ja-JP.md'>日本語\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n👉 AnythingLLM for desktop (Mac, Windows, & Linux)! \u003Ca href=\"https:\u002F\u002Fanythingllm.com\u002Fdownload\" target=\"_blank\"> Download Now\u003C\u002Fa>\n\u003C\u002Fp>\n\nChat with your docs. Automate complex workflows with AI Agents. Hyper-configurable, multi-user ready, battle-tested—and runs locally by default with zero setup friction.\n\n![Chatting](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_3c405f824478.gif)\n\n\u003Cdetails>\n\u003Csummary>\u003Ckbd>Watch the demo!\u003C\u002Fkbd>\u003C\u002Fsummary>\n\n[![Watch the video](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_5b2c33cb0fd3.png)](https:\u002F\u002Fyoutu.be\u002Ff95rGD9trL0)\n\n\u003C\u002Fdetails>\n\n### Product Overview\n\nAnythingLLM is the all-in-one AI application that lets you build a private, fully-featured ChatGPT—without compromises. Connect your favorite local or cloud LLM, ingest your documents, and start chatting in minutes. Out of the box you get built-in agents, multi-user support, vector databases, and document pipelines — no extra configuration required.\n\nAnythingLLM supports multiple users as well where you can control the access and experience per user without compromising the security or privacy of the instance or your intellectual property.\n\n## Cool features of AnythingLLM\n\n- [Intelligent Skill Selection](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent\u002Fintelligent-tool-selection) Enable **unlimited** tools for your models while reducing token usage by up to 80% per query\n- [**No-code AI Agent builder**](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent-flows\u002Foverview)\n- [**Full MCP-compatibility**](https:\u002F\u002Fdocs.anythingllm.com\u002Fmcp-compatibility\u002Foverview)\n- **Multi-modal support (both closed and open-source LLMs!)**\n- [**Custom AI Agents**](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent\u002Fcustom\u002Fintroduction)\n- 👤 Multi-user instance support and permissioning _Docker version only_\n- 🦾 Agents inside your workspace (browse the web, etc)\n- 💬 [Custom Embeddable Chat widget for your website](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanythingllm-embed\u002Fblob\u002Fmain\u002FREADME.md) _Docker version only_\n- 📖 Multiple document type support (PDF, TXT, DOCX, etc)\n- Intuitive chat UI with drag-and-drop uploads and source citations.\n- Production-ready for any cloud deployment.\n- Works with all popular [closed and open-source LLM providers](#supported-llms-embedder-models-speech-models-and-vector-databases).\n- Built-in optimizations for large document sets—lower costs and faster responses than other chat UIs.\n- Full Developer API for custom integrations!\n- ...and much more—install in minutes and see for yourself.\n\n### Supported LLMs, Embedder Models, Speech models, and Vector Databases\n\n**Large Language Models (LLMs):**\n\n- [Any open-source llama.cpp compatible model](\u002Fserver\u002Fstorage\u002Fmodels\u002FREADME.md#text-generation-llm-selection)\n- [OpenAI](https:\u002F\u002Fopenai.com)\n- [OpenAI (Generic)](https:\u002F\u002Fopenai.com)\n- [Azure OpenAI](https:\u002F\u002Fazure.microsoft.com\u002Fen-us\u002Fproducts\u002Fai-services\u002Fopenai-service)\n- [AWS Bedrock](https:\u002F\u002Faws.amazon.com\u002Fbedrock\u002F)\n- [Anthropic](https:\u002F\u002Fwww.anthropic.com\u002F)\n- [NVIDIA NIM (chat models)](https:\u002F\u002Fbuild.nvidia.com\u002Fexplore\u002Fdiscover)\n- [Google Gemini Pro](https:\u002F\u002Fai.google.dev\u002F)\n- [Hugging Face (chat models)](https:\u002F\u002Fhuggingface.co\u002F)\n- [Ollama (chat models)](https:\u002F\u002Follama.ai\u002F)\n- [LM Studio (all models)](https:\u002F\u002Flmstudio.ai)\n- [LocalAI (all models)](https:\u002F\u002Flocalai.io\u002F)\n- [Together AI (chat models)](https:\u002F\u002Fwww.together.ai\u002F)\n- [Fireworks AI (chat models)](https:\u002F\u002Ffireworks.ai\u002F)\n- [Perplexity (chat models)](https:\u002F\u002Fwww.perplexity.ai\u002F)\n- [OpenRouter (chat models)](https:\u002F\u002Fopenrouter.ai\u002F)\n- [DeepSeek (chat models)](https:\u002F\u002Fdeepseek.com\u002F)\n- [Mistral](https:\u002F\u002Fmistral.ai\u002F)\n- [Groq](https:\u002F\u002Fgroq.com\u002F)\n- [Cohere](https:\u002F\u002Fcohere.com\u002F)\n- [KoboldCPP](https:\u002F\u002Fgithub.com\u002FLostRuins\u002Fkoboldcpp)\n- [LiteLLM](https:\u002F\u002Fgithub.com\u002FBerriAI\u002Flitellm)\n- [Text Generation Web UI](https:\u002F\u002Fgithub.com\u002Foobabooga\u002Ftext-generation-webui)\n- [Apipie](https:\u002F\u002Fapipie.ai\u002F)\n- [xAI](https:\u002F\u002Fx.ai\u002F)\n- [Z.AI (chat models)](https:\u002F\u002Fz.ai\u002Fmodel-api)\n- [Novita AI (chat models)](https:\u002F\u002Fnovita.ai\u002Fmodel-api\u002Fproduct\u002Fllm-api?utm_source=github_anything-llm&utm_medium=github_readme&utm_campaign=link)\n- [PPIO](https:\u002F\u002Fppinfra.com?utm_source=github_anything-llm)\n- [Gitee AI](https:\u002F\u002Fai.gitee.com\u002F)\n- [Moonshot AI](https:\u002F\u002Fwww.moonshot.ai\u002F)\n- [Microsoft Foundry Local](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FFoundry-Local)\n- [CometAPI (chat models)](https:\u002F\u002Fapi.cometapi.com\u002F)\n- [Docker Model Runner](https:\u002F\u002Fdocs.docker.com\u002Fai\u002Fmodel-runner\u002F)\n- [PrivateModeAI (chat models)](https:\u002F\u002Fprivatemode.ai\u002F)\n- [SambaNova Cloud (chat models)](https:\u002F\u002Fcloud.sambanova.ai\u002F)\n- [Lemonade by AMD](https:\u002F\u002Flemonade-server.ai)\n\n**Embedder models:**\n\n- [AnythingLLM Native Embedder](\u002Fserver\u002Fstorage\u002Fmodels\u002FREADME.md) (default)\n- [OpenAI](https:\u002F\u002Fopenai.com)\n- [Azure OpenAI](https:\u002F\u002Fazure.microsoft.com\u002Fen-us\u002Fproducts\u002Fai-services\u002Fopenai-service)\n- [LocalAI (all)](https:\u002F\u002Flocalai.io\u002F)\n- [Ollama (all)](https:\u002F\u002Follama.ai\u002F)\n- [LM Studio (all)](https:\u002F\u002Flmstudio.ai)\n- [Cohere](https:\u002F\u002Fcohere.com\u002F)\n\n**Audio Transcription models:**\n\n- [AnythingLLM Built-in](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Ftree\u002Fmaster\u002Fserver\u002Fstorage\u002Fmodels#audiovideo-transcription) (default)\n- [OpenAI](https:\u002F\u002Fopenai.com\u002F)\n\n**TTS (text-to-speech) support:**\n\n- Native Browser Built-in (default)\n- [PiperTTSLocal - runs in browser](https:\u002F\u002Fgithub.com\u002Frhasspy\u002Fpiper)\n- [OpenAI TTS](https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fguides\u002Ftext-to-speech\u002Fvoice-options)\n- [ElevenLabs](https:\u002F\u002Felevenlabs.io\u002F)\n- Any OpenAI Compatible TTS service.\n\n**STT (speech-to-text) support:**\n\n- Native Browser Built-in (default)\n\n**Vector Databases:**\n\n- [LanceDB](https:\u002F\u002Fgithub.com\u002Flancedb\u002Flancedb) (default)\n- [PGVector](https:\u002F\u002Fgithub.com\u002Fpgvector\u002Fpgvector)\n- [Astra DB](https:\u002F\u002Fwww.datastax.com\u002Fproducts\u002Fdatastax-astra)\n- [Pinecone](https:\u002F\u002Fpinecone.io)\n- [Chroma & ChromaCloud](https:\u002F\u002Ftrychroma.com)\n- [Weaviate](https:\u002F\u002Fweaviate.io)\n- [Qdrant](https:\u002F\u002Fqdrant.tech)\n- [Milvus](https:\u002F\u002Fmilvus.io)\n- [Zilliz](https:\u002F\u002Fzilliz.com)\n\n### Technical Overview\n\nThis monorepo consists of six main sections:\n\n- `frontend`: A viteJS + React frontend that you can run to easily create and manage all your content the LLM can use.\n- `server`: A NodeJS express server to handle all the interactions and do all the vectorDB management and LLM interactions.\n- `collector`: NodeJS express server that processes and parses documents from the UI.\n- `docker`: Docker instructions and build process + information for building from source.\n- `embed`: Submodule for generation & creation of the [web embed widget](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanythingllm-embed).\n- `browser-extension`: Submodule for the [chrome browser extension](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanythingllm-extension).\n\n## 🛳 Self-Hosting\n\nMintplex Labs & the community maintain a number of deployment methods, scripts, and templates that you can use to run AnythingLLM locally. Refer to the table below to read how to deploy on your preferred environment or to automatically deploy.\n| Docker | AWS | GCP | Digital Ocean | Render.com |\n|----------------------------------------|----|-----|---------------|------------|\n| [![Deploy on Docker][docker-btn]][docker-deploy] | [![Deploy on AWS][aws-btn]][aws-deploy] | [![Deploy on GCP][gcp-btn]][gcp-deploy] | [![Deploy on DigitalOcean][do-btn]][do-deploy] | [![Deploy on Render.com][render-btn]][render-deploy] |\n\n| Railway                                             | RepoCloud                                                 | Elestio                                             | Northflank                                                   |\n| --------------------------------------------------- | --------------------------------------------------------- | --------------------------------------------------- | ------------------------------------------------------------ |\n| [![Deploy on Railway][railway-btn]][railway-deploy] | [![Deploy on RepoCloud][repocloud-btn]][repocloud-deploy] | [![Deploy on Elestio][elestio-btn]][elestio-deploy] | [![Deploy on Northflank][northflank-btn]][northflank-deploy] |\n\n[or set up a production AnythingLLM instance without Docker →](.\u002FBARE_METAL.md)\n\n## How to setup for development\n\n- `yarn setup` To fill in the required `.env` files you'll need in each of the application sections (from root of repo).\n  - Go fill those out before proceeding. Ensure `server\u002F.env.development` is filled or else things won't work right.\n- `yarn dev:server` To boot the server locally (from root of repo).\n- `yarn dev:frontend` To boot the frontend locally (from root of repo).\n- `yarn dev:collector` To then run the document collector (from root of repo).\n\n[Learn about documents](.\u002Fserver\u002Fstorage\u002Fdocuments\u002FDOCUMENTS.md)\n\n## Telemetry & Privacy\n\nAnythingLLM by Mintplex Labs Inc contains a telemetry feature that collects anonymous usage information.\n\n\u003Cdetails>\n\u003Csummary>\u003Ckbd>More about Telemetry & Privacy for AnythingLLM\u003C\u002Fkbd>\u003C\u002Fsummary>\n\n### Why?\n\nWe use this information to help us understand how AnythingLLM is used, to help us prioritize work on new features and bug fixes, and to help us improve AnythingLLM's performance and stability.\n\n### Opting out\n\nSet `DISABLE_TELEMETRY` in your server or docker .env settings to \"true\" to opt out of telemetry. You can also do this in-app by going to the sidebar > `Privacy` and disabling telemetry.\n\n### What do you explicitly track?\n\nWe will only track usage details that help us make product and roadmap decisions, specifically:\n\n- Type of your installation (Docker or Desktop)\n\n- When a document is added or removed. No information _about_ the document. Just that the event occurred. This gives us an idea of use.\n\n- Type of vector database in use. This helps us prioritize changes when updates arrive for that provider.\n\n- Type of LLM provider & model tag in use. This helps us prioritize changes when updates arrive for that provider or model, or combination thereof. eg: reasoning vs regular, multi-modal models, etc.\n\n- When a chat is sent. This is the most regular \"event\" and gives us an idea of the daily-activity of this project across all installations. Again, only the **event** is sent - we have no information on the nature or content of the chat itself.\n\nYou can verify these claims by finding all locations `Telemetry.sendTelemetry` is called. Additionally these events are written to the output log so you can also see the specific data which was sent - if enabled. **No IP or other identifying information is collected**. The Telemetry provider is [PostHog](https:\u002F\u002Fposthog.com\u002F) - an open-source telemetry collection service.\n\nWe take privacy very seriously, and we hope you understand that we want to learn how our tool is used, without using annoying popup surveys, so we can build something worth using. The anonymous data is _never_ shared with third parties, ever.\n\n[View all telemetry events in source code](https:\u002F\u002Fgithub.com\u002Fsearch?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry(&type=code)\n\n\u003C\u002Fdetails>\n\n## 👋 Contributing\n\n- [Contributing to AnythingLLM](.\u002FCONTRIBUTING.md) - How to contribute to AnythingLLM.\n\n## 💖 Sponsors\n\n### Premium Sponsors\n\n\u003C!-- premium-sponsors (reserved for $100\u002Fmth sponsors who request to be called out here and\u002For are non-private sponsors) -->\n\u003Ca href=\"https:\u002F\u002Fwww.dcsdigital.co.uk\" target=\"_blank\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_0c09536c6d99.png\" height=\"100px\" alt=\"User avatar: DCS DIGITAL\" \u002F>\n\u003C\u002Fa>\n\u003C!-- premium-sponsors -->\n\n### All Sponsors\n\n\u003C!-- all-sponsors -->\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fjaschadub\">\u003Cimg src=\"https://github.com/jaschadub.png\" width=\"60px\" alt=\"User avatar: Jascha\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FKickingAss2024\">\u003Cimg src=\"https://github.com/KickingAss2024.png\" width=\"60px\" alt=\"User avatar: KickAss\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FShadowArcanist\">\u003Cimg src=\"https://github.com/ShadowArcanist.png\" width=\"60px\" alt=\"User avatar: ShadowArcanist\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAtlasVIA\">\u003Cimg src=\"https://github.com/AtlasVIA.png\" width=\"60px\" alt=\"User avatar: Atlas\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fcope\">\u003Cimg src=\"https://github.com/cope.png\" width=\"60px\" alt=\"User avatar: Predrag Stojadinović\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FDiegoSpinola\">\u003Cimg src=\"https://github.com/DiegoSpinola.png\" width=\"60px\" alt=\"User avatar: Diego Spinola\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FPortlandKyGuy\">\u003Cimg src=\"https://github.com/PortlandKyGuy.png\" width=\"60px\" alt=\"User avatar: Kyle\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fpeperunas\">\u003Cimg src=\"https://github.com/peperunas.png\" width=\"60px\" alt=\"User avatar: Giulio De Pasquale\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fjasoncdavis0\">\u003Cimg src=\"https://github.com/jasoncdavis0.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmacstadium\">\u003Cimg src=\"https://github.com/macstadium.png\" width=\"60px\" alt=\"User avatar: MacStadium\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Farmlynobinguar\">\u003Cimg src=\"https://github.com/armlynobinguar.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FMikeHago\">\u003Cimg src=\"https://github.com/MikeHago.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmaaisde\">\u003Cimg src=\"https://github.com/maaisde.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmhollier117\">\u003Cimg src=\"https://github.com/mhollier117.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fpleabargain\">\u003Cimg src=\"https://github.com/pleabargain.png\" width=\"60px\" alt=\"User avatar: Dennis\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fbroichan\">\u003Cimg src=\"https://github.com/broichan.png\" width=\"60px\" alt=\"User avatar: Michael Hamilton, Ph.D.\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fazim-charaniya\">\u003Cimg src=\"https://github.com/azim-charaniya.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fgabriellemon\">\u003Cimg src=\"https://github.com/gabriellemon.png\" width=\"60px\" alt=\"User avatar: TernaryLabs\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FCelaDaniel\">\u003Cimg src=\"https://github.com/CelaDaniel.png\" width=\"60px\" alt=\"User avatar: Daniel Cela\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Faltrsadmin\">\u003Cimg src=\"https://github.com/altrsadmin.png\" width=\"60px\" alt=\"User avatar: Alesso\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fbitjungle\">\u003Cimg src=\"https://github.com/bitjungle.png\" width=\"60px\" alt=\"User avatar: Rune Mathisen\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FpcrossleyAC\">\u003Cimg src=\"https://github.com/pcrossleyAC.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsaroj-pattnaik\">\u003Cimg src=\"https://github.com/saroj-pattnaik.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ftechmedic5\">\u003Cimg src=\"https://github.com/techmedic5.png\" width=\"60px\" alt=\"User avatar: Alan\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fddocta\">\u003Cimg src=\"https://github.com/ddocta.png\" width=\"60px\" alt=\"User avatar: Damien Peters\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fdcsdigital\">\u003Cimg src=\"https://github.com/dcsdigital.png\" width=\"60px\" alt=\"User avatar: DCS Digital\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fpm7y\">\u003Cimg src=\"https://github.com/pm7y.png\" width=\"60px\" alt=\"User avatar: Paul Mcilreavy\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ftilwolf\">\u003Cimg src=\"https://github.com/tilwolf.png\" width=\"60px\" alt=\"User avatar: Til Wolf\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fozzyoss77\">\u003Cimg src=\"https://github.com/ozzyoss77.png\" width=\"60px\" alt=\"User avatar: Leopoldo Crhistian Riverin Gomez\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAlphaEcho11\">\u003Cimg src=\"https://github.com/AlphaEcho11.png\" width=\"60px\" alt=\"User avatar: AJEsau\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsvanomm\">\u003Cimg src=\"https://github.com/svanomm.png\" width=\"60px\" alt=\"User avatar: Steven VanOmmeren\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsocketbox\">\u003Cimg src=\"https://github.com/socketbox.png\" width=\"60px\" alt=\"User avatar: Casey Boettcher\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fzebbern\">\u003Cimg src=\"https://github.com/zebbern.png\" width=\"60px\" alt=\"User avatar: \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Favineetbespin\">\u003Cimg src=\"https://github.com/avineetbespin.png\" width=\"60px\" alt=\"User avatar: Avineet\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Finvictus-1\">\u003Cimg src=\"https://github.com/invictus-1.png\" width=\"60px\" alt=\"User avatar: Chris\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmirbyte\">\u003Cimg src=\"https://github.com/mirbyte.png\" width=\"60px\" alt=\"User avatar: mirko\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fbisonbet\">\u003Cimg src=\"https://github.com/bisonbet.png\" width=\"60px\" alt=\"User avatar: Tim Champ\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FSinkingdev\">\u003Cimg src=\"https://github.com/Sinkingdev.png\" width=\"60px\" alt=\"User avatar: Peter Mathisen\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FEd-STEM\">\u003Cimg src=\"https://github.com/Ed-STEM.png\" width=\"60px\" alt=\"User avatar: Ed di Girolamo\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmilkowski\">\u003Cimg src=\"https://github.com/milkowski.png\" width=\"60px\" alt=\"User avatar: Wojciech Miłkowski\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FADS-Fund\">\u003Cimg src=\"https://github.com/ADS-Fund.png\" width=\"60px\" alt=\"User avatar: ADS Fund\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Farc46-io\">\u003Cimg src=\"https://github.com/arc46-io.png\" width=\"60px\" alt=\"User avatar: arc46 GmbH\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fliyin2015\">\u003Cimg src=\"https://github.com/liyin2015.png\" width=\"60px\" alt=\"User avatar: Li Yin\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FSylphAI-Inc\">\u003Cimg src=\"https://github.com/SylphAI-Inc.png\" width=\"60px\" alt=\"User avatar: SylphAI\" \u002F>\u003C\u002Fa>\u003C!-- all-sponsors -->\n\n## 🌟 Contributors\n\n[![anythingllm contributors](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_0fcb31b2f44b.png)](https:\u002F\u002Fgithub.com\u002Fmintplex-labs\u002Fanything-llm\u002Fgraphs\u002Fcontributors)\n\n[![Star History Chart](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_4d609b6b530d.png)](https:\u002F\u002Fstar-history.com\u002F#mintplex-labs\u002Fanything-llm&Date)\n\n## 🔗 More Products\n\n- **[VectorAdmin][vector-admin]:** An all-in-one GUI & tool-suite for managing vector databases.\n- **[OpenAI Assistant Swarm][assistant-swarm]:** Turn your entire library of OpenAI assistants into one single army commanded from a single agent.\n\n\u003Cdiv align=\"right\">\n\n[![][back-to-top]](#readme-top)\n\n\u003C\u002Fdiv>\n\n---\n\nCopyright © 2026 [Mintplex Labs][profile-link]. \u003Cbr \u002F>\nThis project is [MIT](.\u002FLICENSE) licensed.\n\n\u003C!-- LINK GROUP -->\n\n[back-to-top]: https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-BACK_TO_TOP-222628?style=flat-square\n[profile-link]: https:\u002F\u002Fgithub.com\u002Fmintplex-labs\n[vector-admin]: https:\u002F\u002Fgithub.com\u002Fmintplex-labs\u002Fvector-admin\n[assistant-swarm]: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fopenai-assistant-swarm\n[docker-btn]: .\u002Fimages\u002FdeployBtns\u002Fdocker.png\n[docker-deploy]: .\u002Fdocker\u002FHOW_TO_USE_DOCKER.md\n[aws-btn]: .\u002Fimages\u002FdeployBtns\u002Faws.png\n[aws-deploy]: .\u002Fcloud-deployments\u002Faws\u002Fcloudformation\u002FDEPLOY.md\n[gcp-btn]: https:\u002F\u002Fdeploy.cloud.run\u002Fbutton.svg\n[gcp-deploy]: .\u002Fcloud-deployments\u002Fgcp\u002Fdeployment\u002FDEPLOY.md\n[do-btn]: https:\u002F\u002Fwww.deploytodo.com\u002Fdo-btn-blue.svg\n[do-deploy]: .\u002Fcloud-deployments\u002Fdigitalocean\u002Fterraform\u002FDEPLOY.md\n[render-btn]: https:\u002F\u002Frender.com\u002Fimages\u002Fdeploy-to-render-button.svg\n[render-deploy]: https:\u002F\u002Frender.com\u002Fdeploy?repo=https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm&branch=render\n[render-btn]: https:\u002F\u002Frender.com\u002Fimages\u002Fdeploy-to-render-button.svg\n[render-deploy]: https:\u002F\u002Frender.com\u002Fdeploy?repo=https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm&branch=render\n[railway-btn]: https:\u002F\u002Frailway.app\u002Fbutton.svg\n[railway-deploy]: https:\u002F\u002Frailway.app\u002Ftemplate\u002FHNSCS1?referralCode=WFgJkn\n[repocloud-btn]: https:\u002F\u002Fd16t0pc4846x52.cloudfront.net\u002Fdeploylobe.svg\n[repocloud-deploy]: https:\u002F\u002Frepocloud.io\u002Fdetails\u002F?app_id=276\n[elestio-btn]: https:\u002F\u002Felest.io\u002Fimages\u002Flogos\u002Fdeploy-to-elestio-btn.png\n[elestio-deploy]: https:\u002F\u002Felest.io\u002Fopen-source\u002Fanythingllm\n[northflank-btn]: https:\u002F\u002Fassets.northflank.com\u002Fdeploy_to_northflank_smm_36700fb050.svg\n[northflank-deploy]: https:\u002F\u002Fnorthflank.com\u002Fstacks\u002Fdeploy-anythingllm\n","\u003Ca name=\"readme-top\">\u003C\u002Fa>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fanythingllm.com\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_e38b611303bd.png\" alt=\"AnythingLLM logo\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cdiv align='center'>\n\u003Ca href=\"https:\u002F\u002Ftrendshift.io\u002Frepositories\u002F2415\" target=\"_blank\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_4a68feb902da.png\" alt=\"Mintplex-Labs%2Fanything-llm | Trendshift\" style=\"width: 250px; height: 55px;\" width=\"250\" height=\"55\"\u002F>\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n\u003Cp align=\"center\">\n    \u003Cb>AnythingLLM：\u003C\u002Fb> 你一直在寻找的一站式 AI 应用。\u003Cbr \u002F>\n    与你的文档聊天，使用 AI 代理，高度可配置，支持多用户，无需繁琐的设置。\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002F6UyHPeGZAC\" target=\"_blank\">\n      \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fchat-mintplex_labs-blue.svg?style=flat&logo=data:image\u002Fpng;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAAAIGNIUk0AAHomAACAhAAA+gAAAIDoAAB1MAAA6mAAADqYAAAXcJy6UTwAAAH1UExURQAAAP\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002Fr6+ubn5+7u7\u002F3+\u002Fv39\u002Fenq6urq6\u002Fv7+97f39rb26eoqT1BQ0pOT4+Rkuzs7cnKykZKS0NHSHl8fdzd3ejo6UxPUUBDRdzc3RwgIh8jJSAkJm5xcvHx8aanqB4iJFBTVezt7V5hYlJVVuLj43p9fiImKCMnKZKUlaaoqSElJ21wcfT09O3u7uvr6zE0Nr6\u002FwCUpK5qcnf7+\u002Fnh7fEdKTHx+f0tPUOTl5UZJSx0hIzQ4Odna2zM3Obm7u3x\u002FgKSmp9jZ2T1AQu\u002Fv71pdXkVISr2+vygsLiInKTg7PaOlpisvMcXGxzk8PldaXPLy8u7u7rm6u7S1tsDBwvj4+MPExbe4ueXm5s\u002FQ0Kyf7ewAAAAodFJOUwAABClsrNjx\u002FQM2l9\u002F7lhmI6jTB\u002FkA1GgKJN+nea6vy\u002FMLZQYeVKK3rVA5tAAAAAWJLR0QB\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......\u003Ca name=\"readme-top\">\u003C\u002Fa>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fanythingllm.com\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_e38b611303bd.png\" alt=\"AnythingLLM logo\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cdiv align='center'>\n\u003Ca href=\"https:\u002F\u002Ftrendshift.io\u002Frepositories\u002F2415\" target=\"_blank\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_4a68feb902da.png\" alt=\"Mintplex-Labs%2Fanything-llm | Trendshift\" style=\"width: 250px; height: 55px;\" width=\"250\" height=\"55\"\u002F>\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n\u003Cp align=\"center\">\n    \u003Cb>AnythingLLM:\u003C\u002Fb> 您一直在寻找的一站式AI应用。\u003Cbr \u002F>\n    与您的文档聊天，使用AI智能体，高度可配置，支持多用户，无需繁琐的设置。\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002F6UyHPeGZAC\" target=\"_blank\">\n      \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fchat-mintplex_labs-blue.svg?style=flat&logo=data:image\u002Fpng;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAAAIGNIUk0AAHomAACAhAAA+gAAAIDoAAB1MAAA6mAAADqYAAAXcJy6UTwAAAH1UExURQAAAP\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002F\u002Fr6+ubn5+7u7\u002F3+\u002Fv79\u002Fenq6urq6\u002Fv7+97f39rb26eoqT1BQ0pOT4+Rkuzs7cnKykZKS0NHSHl8fdzd3ejo6UxPUUBDRdzc3RwgIh8jJSAkJm5xcvHx8aanqB4iJFBTVezt7V5hYlJVVuLj43p9fiImKCMnKZKUlaaoqSElJ21wcfT09O3u7uvr6zE0Nr6\u002FwCUpK5qcnf7+\u002Fnh7fEdKTHx+f0tPUOTl5UZJSx0hIzQ4Odna2zM3Obm7u3x\u002FgKSmp9jZ2T1AQu\u002Fv71pdXkVISr2+vygsLiInKTg7PaOlpisvMcXGxzk8PldaXPLy8u7u7rm6u7S1tsDBwvj4+MPExbe4ueXm5s\u002FQ0Kyf7ewAAAAodFJOUwAABClsrNjx\u002FQM2l9\u002F7lhmI6jTB\u002FkA1GgKJN+nea6vy\u002FMLZQYeVKK3rVA5tAAAAAWJLR0QB\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\" alt=\"Discord\">\n  \u003C\u002Fa> |\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fblob\u002Fmaster\u002FLICENSE\" target=\"_blank\">\n      \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fstatic\u002Fv1?label=license&message=MIT&color=white\" alt=\"License\">\n  \u003C\u002Fa> |\n  \u003Ca href=\"https:\u002F\u002Fdocs.anythingllm.com\" target=\"_blank\">\n    文档\n  \u003C\u002Fa> |\n   \u003Ca href=\"https:\u002F\u002Fmy.mintplexlabs.com\u002Faio-checkout?product=anythingllm\" target=\"_blank\">\n    托管实例\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cb>English\u003C\u002Fb> · \u003Ca href='.\u002Flocales\u002FREADME.zh-CN.md'>简体中文\u003C\u002Fa> · \u003Ca href='.\u002Flocales\u002FREADME.ja-JP.md'>日本語\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n👉 AnythingLLM桌面版（Mac、Windows和Linux）！\u003Ca href=\"https:\u002F\u002Fanythingllm.com\u002Fdownload\" target=\"_blank\">立即下载\u003C\u002Fa>\n\u003C\u002Fp>\n\n与您的文档聊天。通过AI智能体自动化复杂的工作流。高度可配置，支持多用户，经过实战考验——默认情况下可在本地运行，无需任何设置即可轻松上手。\n\n![聊天](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_3c405f824478.gif)\n\n\u003Cdetails>\n\u003Csummary>\u003Ckbd>观看演示！\u003C\u002Fkbd>\u003C\u002Fsummary>\n\n[![观看视频](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_5b2c33cb0fd3.png)](https:\u002F\u002Fyoutu.be\u002Ff95rGD9trL0)\n\n\u003C\u002Fdetails>\n\n### 产品概述\n\nAnythingLLM是一款一站式AI应用，让您无需妥协即可构建一个功能齐全的私有ChatGPT。连接您喜爱的本地或云端大模型，导入您的文档，几分钟内即可开始聊天。开箱即用，内置智能体、多用户支持、向量数据库和文档处理管道——无需额外配置。\n\nAnythingLLM还支持多用户模式，您可以为每个用户控制访问权限和使用体验，同时确保实例的安全性以及您的知识产权不受影响。\n\n## AnythingLLM的亮点功能\n\n- [智能工具选择](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent\u002Fintelligent-tool-selection) 为您的模型启用**无限**工具，同时将每次查询的token消耗降低多达80%\n- [**无代码AI智能体构建器**](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent-flows\u002Foverview)\n- [**完全兼容MCP协议**](https:\u002F\u002Fdocs.anythingllm.com\u002Fmcp-compatibility\u002Foverview)\n- **多模态支持（包括闭源和开源大模型！）**\n- [**自定义AI智能体**](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent\u002Fcustom\u002Fintroduction)\n- 👤 多用户实例支持及权限管理 _仅限Docker版本_\n- 🦾 工作空间内的智能体（如浏览网页等）\n- 💬 [适用于您网站的自定义嵌入式聊天组件](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanythingllm-embed\u002Fblob\u002Fmain\u002FREADME.md) _仅限Docker版本_\n- 📖 支持多种文档类型（PDF、TXT、DOCX等）\n- 直观的聊天界面，支持拖放上传和来源引用。\n- 适合任何云环境的生产级部署。\n- 兼容所有主流的[闭源和开源大模型提供商](#supported-llms-embedder-models-speech-models-and-vector-databases)。\n- 针对大型文档集进行了内置优化——成本更低，响应速度更快，优于其他聊天界面。\n- 提供完整的开发者API，方便自定义集成！\n- …还有更多功能——只需几分钟安装，亲身体验吧。\n\n### 支持的 LLM、嵌入模型、语音模型和向量数据库\n\n**大型语言模型 (LLMs)：**\n\n- [任何兼容 llama.cpp 的开源模型](\u002Fserver\u002Fstorage\u002Fmodels\u002FREADME.md#text-generation-llm-selection)\n- [OpenAI](https:\u002F\u002Fopenai.com)\n- [OpenAI（通用）](https:\u002F\u002Fopenai.com)\n- [Azure OpenAI](https:\u002F\u002Fazure.microsoft.com\u002Fen-us\u002Fproducts\u002Fai-services\u002Fopenai-service)\n- [AWS Bedrock](https:\u002F\u002Faws.amazon.com\u002Fbedrock\u002F)\n- [Anthropic](https:\u002F\u002Fwww.anthropic.com\u002F)\n- [NVIDIA NIM（聊天模型）](https:\u002F\u002Fbuild.nvidia.com\u002Fexplore\u002Fdiscover)\n- [Google Gemini Pro](https:\u002F\u002Fai.google.dev\u002F)\n- [Hugging Face（聊天模型）](https:\u002F\u002Fhuggingface.co\u002F)\n- [Ollama（聊天模型）](https:\u002F\u002Follama.ai\u002F)\n- [LM Studio（所有模型）](https:\u002F\u002Flmstudio.ai)\n- [LocalAI（所有模型）](https:\u002F\u002Flocalai.io\u002F)\n- [Together AI（聊天模型）](https:\u002F\u002Fwww.together.ai\u002F)\n- [Fireworks AI（聊天模型）](https:\u002F\u002Ffireworks.ai\u002F)\n- [Perplexity（聊天模型）](https:\u002F\u002Fwww.perplexity.ai\u002F)\n- [OpenRouter（聊天模型）](https:\u002F\u002Fopenrouter.ai\u002F)\n- [DeepSeek（聊天模型）](https:\u002F\u002Fdeepseek.com\u002F)\n- [Mistral](https:\u002F\u002Fmistral.ai\u002F)\n- [Groq](https:\u002F\u002Fgroq.com\u002F)\n- [Cohere](https:\u002F\u002Fcohere.com\u002F)\n- [KoboldCPP](https:\u002F\u002Fgithub.com\u002FLostRuins\u002Fkoboldcpp)\n- [LiteLLM](https:\u002F\u002Fgithub.com\u002FBerriAI\u002Flitellm)\n- [Text Generation Web UI](https:\u002F\u002Fgithub.com\u002Foobabooga\u002Ftext-generation-webui)\n- [Apipie](https:\u002F\u002Fapipie.ai\u002F)\n- [xAI](https:\u002F\u002Fx.ai\u002F)\n- [Z.AI（聊天模型）](https:\u002F\u002Fz.ai\u002Fmodel-api)\n- [Novita AI（聊天模型）](https:\u002F\u002Fnovita.ai\u002Fmodel-api\u002Fproduct\u002Fllm-api?utm_source=github_anything-llm&utm_medium=github_readme&utm_campaign=link)\n- [PPIO](https:\u002F\u002Fppinfra.com?utm_source=github_anything-llm)\n- [Gitee AI](https:\u002F\u002Fai.gitee.com\u002F)\n- [Moonshot AI](https:\u002F\u002Fwww.moonshot.ai\u002F)\n- [Microsoft Foundry Local](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FFoundry-Local)\n- [CometAPI（聊天模型）](https:\u002F\u002Fapi.cometapi.com\u002F)\n- [Docker Model Runner](https:\u002F\u002Fdocs.docker.com\u002Fai\u002Fmodel-runner\u002F)\n- [PrivateModeAI（聊天模型）](https:\u002F\u002Fprivatemode.ai\u002F)\n- [SambaNova Cloud（聊天模型）](https:\u002F\u002Fcloud.sambanova.ai\u002F)\n- [Lemonade by AMD](https:\u002F\u002Flemonade-server.ai)\n\n**嵌入模型：**\n\n- [AnythingLLM 原生嵌入器](\u002Fserver\u002Fstorage\u002Fmodels\u002FREADME.md)（默认）\n- [OpenAI](https:\u002F\u002Fopenai.com)\n- [Azure OpenAI](https:\u002F\u002Fazure.microsoft.com\u002Fen-us\u002Fproducts\u002Fai-services\u002Fopenai-service)\n- [LocalAI（全部）](https:\u002F\u002Flocalai.io\u002F)\n- [Ollama（全部）](https:\u002F\u002Follama.ai\u002F)\n- [LM Studio（全部）](https:\u002F\u002Flmstudio.ai)\n- [Cohere](https:\u002F\u002Fcohere.com\u002F)\n\n**音频转录模型：**\n\n- [AnythingLLM 内置](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Ftree\u002Fmaster\u002Fserver\u002Fstorage\u002Fmodels#audiovideo-transcription)（默认）\n- [OpenAI](https:\u002F\u002Fopenai.com\u002F)\n\n**TTS（文本到语音）支持：**\n\n- 浏览器原生内置（默认）\n- [PiperTTSLocal — 在浏览器中运行](https:\u002F\u002Fgithub.com\u002Frhasspy\u002Fpiper)\n- [OpenAI TTS](https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fguides\u002Ftext-to-speech\u002Fvoice-options)\n- [ElevenLabs](https:\u002F\u002Felevenlabs.io\u002F)\n- 任何兼容 OpenAI 的 TTS 服务。\n\n**STT（语音到文本）支持：**\n\n- 浏览器原生内置（默认）\n\n**向量数据库：**\n\n- [LanceDB](https:\u002F\u002Fgithub.com\u002Flancedb\u002Flancedb)（默认）\n- [PGVector](https:\u002F\u002Fgithub.com\u002Fpgvector\u002Fpgvector)\n- [Astra DB](https:\u002F\u002Fwww.datastax.com\u002Fproducts\u002Fdatastax-astra)\n- [Pinecone](https:\u002F\u002Fpinecone.io)\n- [Chroma & ChromaCloud](https:\u002F\u002Ftrychroma.com)\n- [Weaviate](https:\u002F\u002Fweaviate.io)\n- [Qdrant](https:\u002F\u002Fqdrant.tech)\n- [Milvus](https:\u002F\u002Fmilvus.io)\n- [Zilliz](https:\u002F\u002Fzilliz.com)\n\n### 技术概览\n\n这个单体仓库由六个主要部分组成：\n\n- `frontend`：一个基于 viteJS 和 React 的前端，您可以运行它来轻松创建和管理 LLM 可以使用的所有内容。\n- `server`：一个 NodeJS express 服务器，用于处理所有交互，并进行向量数据库管理和 LLM 交互。\n- `collector`：一个 NodeJS express 服务器，用于处理和解析来自 UI 的文档。\n- `docker`：Docker 指令和构建流程，以及从源代码构建的信息。\n- `embed`：用于生成和创建 [网页嵌入小部件](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanythingllm-embed) 的子模块。\n- `browser-extension`：用于 [Chrome 浏览器扩展](https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanythingllm-extension) 的子模块。\n\n## 🛳 自托管\n\nMintplex Labs 和社区维护着多种部署方法、脚本和模板，您可以使用它们在本地运行 AnythingLLM。请参阅下表，了解如何在您首选的环境中部署，或进行自动部署。\n| Docker | AWS | GCP | Digital Ocean | Render.com |\n|----------------------------------------|----|-----|---------------|------------|\n| [![在 Docker 上部署][docker-btn]][docker-deploy] | [![在 AWS 上部署][aws-btn]][aws-deploy] | [![在 GCP 上部署][gcp-btn]][gcp-deploy] | [![在 DigitalOcean 上部署][do-btn]][do-deploy] | [![在 Render.com 上部署][render-btn]][render-deploy] |\n\n| Railway                                             | RepoCloud                                                 | Elestio                                             | Northflank                                                   |\n| --------------------------------------------------- | --------------------------------------------------------- | --------------------------------------------------- | ------------------------------------------------------------ |\n| [![在 Railway 上部署][railway-btn]][railway-deploy] | [![在 RepoCloud 上部署][repocloud-btn]][repocloud-deploy] | [![在 Elestio 上部署][elestio-btn]][elestio-deploy] | [![在 Northflank 上部署][northflank-btn]][northflank-deploy] |\n\n[或者无需 Docker 即可搭建生产级 AnythingLLM 实例 →](.\u002FBARE_METAL.md)\n\n## 开发环境设置方法\n\n- `yarn setup`：用于填充应用程序各部分所需的 `.env` 文件（从仓库根目录开始）。\n  - 请先填写这些文件再继续操作。确保 `server\u002F.env.development` 已填写，否则系统将无法正常工作。\n- `yarn dev:server`：在本地启动服务器（从仓库根目录开始）。\n- `yarn dev:frontend`：在本地启动前端（从仓库根目录开始）。\n- `yarn dev:collector`：随后运行文档收集器（从仓库根目录开始）。\n\n[了解文档相关信息](.\u002Fserver\u002Fstorage\u002Fdocuments\u002FDOCUMENTS.md)\n\n## 遥测与隐私\n\nMintplex Labs Inc. 的 AnythingLLM 包含一项遥测功能，用于收集匿名使用信息。\n\n\u003Cdetails>\n\u003Csummary>\u003Ckbd>关于 AnythingLLM 的遥测与隐私更多信息\u003C\u002Fkbd>\u003C\u002Fsummary>\n\n### 为什么？\n\n我们使用这些信息来帮助我们了解 AnythingLLM 的使用情况，以便优先处理新功能和错误修复，并提升 AnythingLLM 的性能和稳定性。\n\n### 退出选项\n\n在您的服务器或 Docker 的 `.env` 设置中，将 `DISABLE_TELEMETRY` 设置为“true”，即可退出遥测功能。您也可以在应用内通过侧边栏 > `Privacy` 并关闭遥测功能来实现退出。\n\n### 您明确跟踪哪些内容？\n\n我们仅会跟踪有助于我们做出产品和路线图决策的使用详情，具体包括：\n\n- 您的安装类型（Docker 或 Desktop）\n\n- 文档被添加或移除的时间。我们不会收集关于文档本身的任何信息，只记录事件的发生。这能帮助我们了解产品的使用情况。\n\n- 当前使用的向量数据库类型。这有助于我们在该提供商推出更新时确定功能优先级。\n\n- 当前使用的 LLM 提供商及模型标签。这有助于我们在该提供商或特定模型、或两者的组合推出更新时确定功能优先级，例如推理型与常规型、多模态模型等。\n\n- 聊天消息发送的时间。这是最频繁发生的“事件”，能够帮助我们了解该项目在所有安装中的日常活动情况。同样地，我们仅发送**事件**本身——不会收集聊天内容或性质的相关信息。\n\n您可以通过查找所有调用 `Telemetry.sendTelemetry` 的位置来验证这些说法。此外，这些事件也会写入输出日志中，因此如果您启用了日志记录功能，还可以查看实际发送的具体数据。**我们绝不会收集 IP 地址或其他可识别身份的信息**。我们的遥测服务提供商是 [PostHog](https:\u002F\u002Fposthog.com\u002F)——一个开源的遥测数据收集服务。\n\n我们非常重视隐私保护，并希望您能理解，我们希望通过匿名方式了解工具的使用情况，而无需烦人的弹窗调查，从而打造一款真正值得使用的工具。这些匿名数据将**绝不**与第三方共享。\n\n[在源代码中查看所有遥测事件](https:\u002F\u002Fgithub.com\u002Fsearch?q=repo%3AMintplex-Labs%2Fanything-llm%20.sendTelemetry(&type=code)\n\n\u003C\u002Fdetails>\n\n## 👋 参与贡献\n\n- [参与 AnythingLLM 的贡献](.\u002FCONTRIBUTING.md) —— 如何为 AnythingLLM 做出贡献。\n\n## 💖 赞助商\n\n### 高级赞助商\n\n\u003C!-- premium-sponsors (reserved for $100\u002Fmth sponsors who request to be called out here and\u002For are non-private sponsors) -->\n\u003Ca href=\"https:\u002F\u002Fwww.dcsdigital.co.uk\" target=\"_blank\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_0c09536c6d99.png\" height=\"100px\" alt=\"用户头像：DCS DIGITAL\" \u002F>\n\u003C\u002Fa>\n\u003C!-- premium-sponsors -->\n\n### 所有赞助者\n\n\u003C!-- all-sponsors -->\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fjaschadub\">\u003Cimg src=\"https://github.com/jaschadub.png\" width=\"60px\" alt=\"用户头像：Jascha\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FKickingAss2024\">\u003Cimg src=\"https://github.com/KickingAss2024.png\" width=\"60px\" alt=\"用户头像：KickAss\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FShadowArcanist\">\u003Cimg src=\"https://github.com/ShadowArcanist.png\" width=\"60px\" alt=\"用户头像：ShadowArcanist\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAtlasVIA\">\u003Cimg src=\"https://github.com/AtlasVIA.png\" width=\"60px\" alt=\" 用户头像：Atlas\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fcope\">\u003Cimg src=\"https://github.com/cope.png\" width=\"60px\" alt=\"用户头像：Predrag Stojadinović\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FDiegoSpinola\">\u003Cimg src=\"https://github.com/DiegoSpinola.png\" width=\"60px\" alt=\" 用户头像：Diego Spinola\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FPortlandKyGuy\">\u003Cimg src=\"https://github.com/PortlandKyGuy.png\" width=\"60px\" alt=\" 用户头像：Kyle\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fpeperunas\">\u003Cimg src=\"https://github.com/peperunas.png\" width=\"60px\" alt=\" 用户头像：Giulio De Pasquale\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fjasoncdavis0\">\u003Cimg src=\"https://github.com/jasoncdavis0.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmacstadium\">\u003Cimg src=\"https://github.com/macstadium.png\" width=\"60px\" alt=\" 用户头像：MacStadium\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Farmlynobinguar\">\u003Cimg src=\"https://github.com/armlynobinguar.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FMikeHago\">\u003Cimg src=\"https://github.com/MikeHago.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmaaisde\">\u003Cimg src=\"https://github.com/maaisde.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmhollier117\">\u003Cimg src=\"https://github.com/mhollier117.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fpleabargain\">\u003Cimg src=\"https://github.com/pleabargain.png\" width=\"60px\" alt=\" 用户头像：Dennis\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fbroichan\">\u003Cimg src=\"https://github.com/broichan.png\" width=\"60px\" alt=\" 用户头像：Michael Hamilton, Ph.D.\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fazim-charaniya\">\u003Cimg src=\"https://github.com/azim-charaniya.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fgabriellemon\">\u003Cimg src=\"https://github.com/gabriellemon.png\" width=\"60px\" alt=\" 用户头像：TernaryLabs\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FCelaDaniel\">\u003Cimg src=\"https://github.com/CelaDaniel.png\" width=\"60px\" alt=\" 用户头像：Daniel Cela\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Faltrsadmin\">\u003Cimg src=\"https://github.com/altrsadmin.png\" width=\"60px\" alt=\" 用户头像：Alesso\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fbitjungle\">\u003Cimg src=\"https://github.com/bitjungle.png\" width=\"60px\" alt=\" 用户头像：Rune Mathisen\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FpcrossleyAC\">\u003Cimg src=\"https://github.com/pcrossleyAC.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsaroj-pattnaik\">\u003Cimg src=\"https://github.com/saroj-pattnaik.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ftechmedic5\">\u003Cimg src=\"https://github.com/techmedic5.png\" width=\"60px\" alt=\" 用户头像：Alan\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fddocta\">\u003Cimg src=\"https://github.com/ddocta.png\" width=\"60px\" alt=\" 用户头像：Damien Peters\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fdcsdigital\">\u003Cimg src=\"https://github.com/dcsdigital.png\" width=\"60px\" alt=\" 用户头像：DCS Digital\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fpm7y\">\u003Cimg src=\"https://github.com/pm7y.png\" width=\"60px\" alt=\" 用户头像：Paul Mcilreavy\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ftilwolf\">\u003Cimg src=\"https://github.com/tilwolf.png\" width=\"60px\" alt=\" 用户头像：Til Wolf\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fozzyoss77\">\u003Cimg src=\"https://github.com/ozzyoss77.png\" width=\"60px\" alt=\" 用户头像：Leopoldo Crhistian Riverin Gomez\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FAlphaEcho11\">\u003Cimg src=\"https://github.com/AlphaEcho11.png\" width=\"60px\" alt=\" 用户头像：AJEsau\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsvanomm\">\u003Cimg src=\"https://github.com/svanomm.png\" width=\"60px\" alt=\" 用户头像：Steven VanOmmeren\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsocketbox\">\u003Cimg src=\"https://github.com/socketbox.png\" width=\"60px\" alt=\" 用户头像：Casey Boettcher\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fzebbern\">\u003Cimg src=\"https://github.com/zebbern.png\" width=\"60px\" alt=\" 用户头像： \" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Favineetbespin\">\u003Cimg src=\"https://github.com/avineetbespin.png\" width=\"60px\" alt=\" 用户头像：Avineet\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Finvictus-1\">\u003Cimg src=\"https://github.com/invictus-1.png\" width=\"60px\" alt=\" 用户头像：Chris\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmirbyte\">\u003Cimg src=\"https://github.com/mirbyte.png\" width=\"60px\" alt=\" 用户头像：mirko\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fbisonbet\">\u003Cimg src=\"https://github.com/bisonbet.png\" width=\"60px\" alt=\" 用户头像：Tim Champ\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FSinkingdev\">\u003Cimg src=\"https://github.com/Sinkingdev.png\" width=\"60px\" alt=\" 用户头像：Peter Mathisen\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FEd-STEM\">\u003Cimg src=\"https://github.com/Ed-STEM.png\" width=\"60px\" alt=\" 用户头像：Ed di Girolamo\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmilkowski\">\u003Cimg src=\"https://github.com/milkowski.png\" width=\"60px\" alt=\" 用户头像：Wojciech Miłkowski\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FADS-Fund\">\u003Cimg src=\"https://github.com/ADS-Fund.png\" width=\"60px\" alt=\" 用户头像：ADS Fund\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Farc46-io\">\u003Cimg src=\"https://github.com/arc46-io.png\" width=\"60px\" alt=\" 用户头像：arc46 GmbH\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fliyin2015\">\u003Cimg src=\"https://github.com/liyin2015.png\" width=\"60px\" alt=\" 用户头像：Li Yin\" \u002F>\u003C\u002Fa>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FSylphAI-Inc\">\u003Cimg src=\"https://github.com/SylphAI-Inc.png\" width=\"60px\" alt=\" 用户头像：SylphAI\" \u002F>\u003C\u002Fa>\u003C!-- all-sponsors -->\n\n## 🌟 贡献者\n\n[![anythingllm贡献者](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_0fcb31b2f44b.png)](https:\u002F\u002Fgithub.com\u002Fmintplex-labs\u002Fanything-llm\u002Fgraphs\u002Fcontributors)\n\n[![星标历史图表](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_readme_4d609b6b530d.png)](https:\u002F\u002Fstar-history.com\u002F#mintplex-labs\u002Fanything-llm&Date)\n\n## 🔗 More Products\n\n- **[VectorAdmin][vector-admin]:** An all-in-one GUI & tool-suite for managing vector databases.\n- **[OpenAI Assistant Swarm][assistant-swarm]:** Turn your entire library of OpenAI assistants into one single army commanded from a single agent.\n\n\u003Cdiv align=\"right\">\n\n[![][back-to-top]](#readme-top)\n\n\u003C\u002Fdiv>\n\n---\n\nCopyright © 2026 [Mintplex Labs][profile-link]. \u003Cbr \u002F>\nThis project is [MIT](.\u002FLICENSE) licensed.\n\n\u003C!-- LINK GROUP -->\n\n[back-to-top]: https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-BACK_TO_TOP-222628?style=flat-square\n[profile-link]: https:\u002F\u002Fgithub.com\u002Fmintplex-labs\n[vector-admin]: https:\u002F\u002Fgithub.com\u002Fmintplex-labs\u002Fvector-admin\n[assistant-swarm]: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fopenai-assistant-swarm\n[docker-btn]: .\u002Fimages\u002FdeployBtns\u002Fdocker.png\n[docker-deploy]: .\u002Fdocker\u002FHOW_TO_USE_DOCKER.md\n[aws-btn]: .\u002Fimages\u002FdeployBtns\u002Faws.png\n[aws-deploy]: .\u002Fcloud-deployments\u002Faws\u002Fcloudformation\u002FDEPLOY.md\n[gcp-btn]: https:\u002F\u002Fdeploy.cloud.run\u002Fbutton.svg\n[gcp-deploy]: .\u002Fcloud-deployments\u002Fgcp\u002Fdeployment\u002FDEPLOY.md\n[do-btn]: https:\u002F\u002Fwww.deploytodo.com\u002Fdo-btn-blue.svg\n[do-deploy]: .\u002Fcloud-deployments\u002Fdigitalocean\u002Fterraform\u002FDEPLOY.md\n[render-btn]: https:\u002F\u002Frender.com\u002Fimages\u002Fdeploy-to-render-button.svg\n[render-deploy]: https:\u002F\u002Frender.com\u002Fdeploy?repo=https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm&branch=render\n[render-btn]: https:\u002F\u002Frender.com\u002Fimages\u002Fdeploy-to-render-button.svg\n[render-deploy]: https:\u002F\u002Frender.com\u002Fdeploy?repo=https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm&branch=render\n[railway-btn]: https:\u002F\u002Frailway.app\u002Fbutton.svg\n[railway-deploy]: https:\u002F\u002Frailway.app\u002Ftemplate\u002FHNSCS1?referralCode=WFgJkn\n[repocloud-btn]: https:\u002F\u002Fd16t0pc4846x52.cloudfront.net\u002Fdeploylobe.svg\n[repocloud-deploy]: https:\u002F\u002Frepocloud.io\u002Fdetails\u002F?app_id=276\n[elestio-btn]: https:\u002F\u002Felest.io\u002Fimages\u002Flogos\u002Fdeploy-to-elestio-btn.png\n[elestio-deploy]: https:\u002F\u002Felest.io\u002Fopen-source\u002Fanythingllm\n[northflank-btn]: https:\u002F\u002Fassets.northflank.com\u002Fdeploy_to_northflank_smm_36700fb050.svg\n[northflank-deploy]: https:\u002F\u002Fnorthflank.com\u002Fstacks\u002Fdeploy-anythingllm","# AnythingLLM 快速上手指南\n\nAnythingLLM 是一款全能型 AI 应用，支持本地部署。你可以连接本地或云端大模型，导入文档进行对话，构建无代码 AI Agent，并支持多用户管理。默认情况下可本地运行，无需复杂配置。\n\n## 环境准备\n\n### 系统要求\n- **操作系统**：Windows, macOS, Linux\n- **内存**：建议 8GB 以上（运行本地模型需更多）\n- **磁盘空间**：至少 10GB 可用空间\n- **运行时环境**：\n  - Node.js (v18+)\n  - Yarn 包管理器\n  - Docker (可选，用于容器化部署)\n\n### 前置依赖\n确保已安装以下工具：\n```bash\n# 检查 Node.js 版本\nnode -v\n\n# 检查 Yarn 是否安装\nyarn -v\n\n# 若未安装 Yarn，可通过 npm 安装\nnpm install -g yarn\n```\n\n## 安装步骤\n\n### 方式一：桌面客户端（推荐新手）\n直接下载预编译的桌面应用程序，零配置启动：\n- 访问官网下载页：https:\u002F\u002Fanythingllm.com\u002Fdownload\n- 选择对应系统（Mac\u002FWindows\u002FLinux）安装包\n- 安装后直接运行即可\n\n### 方式二：源码部署（开发者模式）\n\n1. **克隆仓库**\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm.git\ncd anything-llm\n```\n\n2. **初始化环境变量**\n```bash\nyarn setup\n```\n> 执行后请根据提示填写 `.env` 文件，特别是 `server\u002F.env.development` 必须配置正确。\n\n3. **启动服务**\n在根目录下分别启动三个核心组件：\n```bash\n# 启动后端服务\nyarn dev:server\n\n# 启动前端界面（新终端窗口）\nyarn dev:frontend\n\n# 启动文档收集器（新终端窗口）\nyarn dev:collector\n```\n\n### 方式三：Docker 部署（生产环境\u002F多用户）\n使用官方提供的 Docker 镜像快速部署（支持多用户和权限管理）：\n```bash\ndocker run -d -p 3000:3000 mintplexlabs\u002Fanythingllm\n```\n*注：具体 Docker 编排文件请参考项目 `docker` 目录或官方文档。*\n\n## 基本使用\n\n1. **配置模型提供商**\n   - 打开浏览器访问 `http:\u002F\u002Flocalhost:3000` (默认端口)。\n   - 在设置中选择支持的 LLM 提供商（如 Ollama, OpenAI, LocalAI 等）。\n   - 填入对应的 API Key 或本地模型路径。\n\n2. **创建知识库工作区**\n   - 点击 \"New Workspace\" 创建一个新的工作空间。\n   - 将文档（PDF, TXT, DOCX 等）拖拽上传至工作区。\n   - 系统会自动解析文档并向量化存储（默认使用 LanceDB）。\n\n3. **开始对话**\n   - 在聊天窗口输入问题，AI 将基于你上传的文档内容进行回答。\n   - 回复中会附带来源引用，方便追溯原文。\n\n4. **进阶功能**\n   - **AI Agent**：在工作区内启用 Agent 功能，让 AI 自主浏览网页或执行复杂任务。\n   - **多模态支持**：上传包含图片或音频的文件进行交互。\n   - **嵌入网站**：(Docker 版) 生成 Embed 代码将聊天窗口嵌入你自己的网站。","某中型科技公司的法务团队需要快速从数百份历史合同和内部合规文档中检索特定条款，以应对突发的审计需求。\n\n### 没有 anything-llm 时\n- **数据隐私风险高**：团队不敢将敏感合同上传至公共 AI 平台，担心商业机密泄露，只能放弃使用高效的 AI 辅助。\n- **检索效率极低**：成员不得不手动打开几十个 PDF 文件逐个搜索关键词，耗时数小时且容易遗漏关键信息。\n- **部署门槛过高**：尝试搭建本地私有化大模型时，被复杂的环境配置、依赖安装和 GPU 调试劝退，项目迟迟无法落地。\n- **协作困难**：缺乏统一的多用户入口，每个人都在自己的电脑上维护独立的脚本，知识无法在团队内共享。\n\n### 使用 anything-llm 后\n- **原生隐私保护**：anything-llm 默认在本地设备运行，所有合同数据不出内网，法务团队可放心地让 AI 处理核心机密文档。\n- **秒级智能问答**：将所有文档拖入 anything-llm 构建知识库后，团队成员可直接用自然语言提问（如“找出所有关于违约赔偿的条款”），瞬间获得精准答案及出处。\n- **零摩擦启动**：无需任何复杂配置或编写代码，下载桌面端即可直接使用，非技术背景的法务人员也能在 5 分钟内完成部署。\n- **无缝团队协作**：anything-llm 支持多用户管理，团队可共享同一个合同知识库，统一对外的回答标准，大幅提升协作效率。\n\nanything-llm 通过“零配置本地部署”与“文档对话”能力，让企业在确保数据绝对隐私的前提下，瞬间将静态文档库转化为可交互的智能知识引擎。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FMintplex-Labs_anything-llm_5b2c33cb.png","Mintplex-Labs","Mintplex Labs","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FMintplex-Labs_2d73495c.png","AI applications that are useful.",null,"team@mintplexlabs.com","mintplexlabs","https:\u002F\u002Fmintplexlabs.com","https:\u002F\u002Fgithub.com\u002FMintplex-Labs",[82,86,90,94,97,101,104],{"name":83,"color":84,"percentage":85},"JavaScript","#f1e05a",98.6,{"name":87,"color":88,"percentage":89},"CSS","#663399",1.1,{"name":91,"color":92,"percentage":93},"Dockerfile","#384d54",0.1,{"name":95,"color":96,"percentage":93},"HTML","#e34c26",{"name":98,"color":99,"percentage":100},"Shell","#89e051",0,{"name":102,"color":103,"percentage":100},"Go Template","#00ADD8",{"name":105,"color":106,"percentage":100},"HCL","#844FBA",58383,6318,"2026-04-15T20:50:12","MIT","Linux, macOS, Windows","非必需（支持纯 CPU 运行或连接外部 LLM 服务）；若本地运行开源模型，需求取决于具体模型大小","未说明（取决于是否本地运行大模型及文档处理量）",{"notes":115,"python":116,"dependencies":117},"该工具主要基于 Node.js 构建，默认本地运行无需复杂配置。支持多种部署方式：桌面客户端（Mac\u002FWindows\u002FLinux）、Docker 容器（支持多用户和嵌入组件）或裸机部署。若需本地运行大语言模型，需额外配置 Ollama、LM Studio 或 llama.cpp 等推理后端，此时硬件需求由所选模型决定。","未说明（核心后端基于 Node.js\u002FExpress，非 Python 应用）",[118,119,120],"Node.js","Yarn","Docker (可选，用于多用户部署)",[52,14,35,16,13,122],"其他",[124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142],"rag","lmstudio","localai","vector-database","ollama","local-llm","llama3","llm","ai-agents","multimodal","custom-ai-agents","deepseek","mcp","mcp-servers","no-code","qwen3","web-scraping","kimi","moonshot","2026-03-27T02:49:30.150509","2026-04-16T16:03:28.195355",[146,151,156,161,166,170],{"id":147,"question_zh":148,"answer_zh":149,"source_url":150},35832,"在 Snapdragon X\u002FNPU 设备上使用 QNN 模型时出现\"QNN Engine is offline\"错误怎么办？","应用程序启动时通常需要 10-20 秒将模型加载到 NPU 共享内存中，此消息通常会自动消失。如果错误持续存在，可能是因为设备的“能源选项”设置导致加载速度过慢。请尝试将电源计划设置为“最佳性能（Best Performance）”，以加快 NPU 的加载速度。","https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fissues\u002F2962",{"id":152,"question_zh":153,"answer_zh":154,"source_url":155},35833,"在 Linux 上使用 AppImage 安装 AnythingLLM 时，创建工作空间报错\"Prisma Client could not locate the Query Engine\"或 OpenSSL 版本不匹配如何解决？","这是由于 AppImage 挂载创建的临时目录是只写的，阻碍了 Prisma 引擎运行数据库迁移。目前有两种解决方案：\n1. 使用官方提供的安装脚本进行非捆绑安装（参考文档：https:\u002F\u002Fdocs.useanything.com\u002Finstallation\u002Fdesktop\u002Flinux）。\n2. 直接在 Linux 上使用 Docker 部署，避免 AppImage 的文件系统限制。","https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fissues\u002F898",{"id":157,"question_zh":158,"answer_zh":159,"source_url":160},35834,"Docker 容器启动时报错\"Illegal instruction (core dumped)\"或退出代码 115 是什么原因？","这通常是因为宿主机的 CPU 不支持 AVX2 指令集（常见于某些 VPS 环境）。AnythingLLM 默认构建需要 AVX2 支持。\n解决方案：\n1. 确认您的 CPU 是否支持 AVX2。\n2. 如果不支持，请拉取不包含该依赖的旧版本镜像：`docker pull mintplexlabs\u002Fanythingllm:lancedb_revert` 并使用该标签运行。","https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fissues\u002F1331",{"id":162,"question_zh":163,"answer_zh":164,"source_url":165},35835,"在 Kubernetes (k8s) 中部署 AnythingLLM 时，应该暴露哪些端口？","在任何 Kubernetes 模板中，唯一需要公开暴露的端口是 **3001**（应用主服务端口）。\n注意事项：\n- 不要暴露 8888 端口（内部健康检查用）。\n- 9090 端口仅用于 Prometheus 指标导出，如果您没有配置相关的 Exporter 或不需要监控指标，也应移除该端口的暴露和相关注解。","https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fissues\u002F1463",{"id":167,"question_zh":168,"answer_zh":169,"source_url":165},35836,"如何在 Kubernetes 中自定义 AnythingLLM 的服务端口？","您可以使用非标准端口（例如 3002）来避免与节点上其他服务（如默认的 3001 端口）冲突。只需在 Deployment 或 Service 配置中修改容器端口映射，并通过环境变量调整配置即可正常工作。同时，请确保移除模板中不必要的健康检查（端口 8888）和指标端口（9090），除非您明确需要它们。",{"id":171,"question_zh":172,"answer_zh":173,"source_url":155},35837,"Linux 用户遇到 AppImage 无法创建 workspace 的问题，除了重装还有什么办法？","这是已知问题，主要原因是 AppImage 的临时文件系统权限限制。除了等待修复外，最稳定的替代方案是直接使用 Docker 运行 AnythingLLM，因为 Linux 原生支持 Docker，可以完全绕过 AppImage 的文件系统挂载问题。",[175,180,185,190,195,200,205,210,215,220,225,230,235,240,245,250,255,260,265,270],{"id":176,"version":177,"summary_zh":178,"released_at":179},281027,"v1.12.0","## 主要功能\n\n\n### 原生工具调用的自动模式\n\n对于支持原生工具调用的[精选提供商](https:\u002F\u002Fdocs.anythingllm.com\u002Ffeatures\u002Fchat-modes#available-chat-modes)，您不再需要使用`@agent`来调用工具。现在，您可以直接使用工具，无需额外指令。\n\n如果您的提示输入中没有“@”符号，聊天将根据需要自动使用工具。\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2ab6af9a-98d0-4d52-a3d3-849d70e64004\n\n## 智能工具选择\n\n我们新增了一项名为[智能工具选择](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent\u002Fintelligent-tool-selection)的功能。此功能允许您为代理加载**无限数量**的工具，并将其纳入上下文中，从而提升性能并使每次对话的令牌使用量最多节省80%。\n\n\u003Cimg width=\"1713\" height=\"898\" alt=\"settings-menu-icon-location\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F06d2eb83-6eff-4ab2-92ba-43ecda5e7523\" \u002F>\n\n## 文件系统代理\n\n我们新增了一项名为[文件系统代理](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent\u002Fusage#file-system-agent)的功能。此功能使您能够使用宿主机的文件系统来搜索文件和目录。\n\n\u003Cimg width=\"3456\" height=\"1810\" alt=\"fs-main-panel\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F142dd8c3-ebb4-4fb9-81dd-ede225c93227\" \u002F>\n\n## 文档生成代理\n\n我们为[文档生成](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent\u002Fusage#document-generation-agent)添加了一个新的内置代理。借助文档生成功能，您可以生成文本文件、PDF、Excel文件、Docx，甚至完整的PowerPoint演示文稿。\n\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F493e27bb-4944-4a8d-8995-a7c8b3eaadd5\n\n\n## Telegram 机器人\n\nAnythingLLM Docker 和 Desktop 现在支持[Telegram 机器人](https:\u002F\u002Fdocs.anythingllm.com\u002Fchannels\u002Ftelegram)，让您可以在世界任何地方连接到您的 AnythingLLM 实例。\n\n**支持功能**：\n- 文本聊天（流式传输与思考中状态）\n- 图像理解\n- 语音消息与附件\n- 自动模式及 @agent 支持\n- 工作空间与线程选择\n- 模型选择\n- 引用\n- AnythingLLM 中所有可用的代理技能\n\n\u003Cimg width=\"400\" height=\"800\" alt=\"image-understanding\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2208bfda-1385-44ef-af1e-704da5c4be25\" \u002F>\n\n## 变更内容\n* 更新 exa 搜索提供商描述，由 @theishangoswami 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5225 中完成\n* 工作空间的自动模式（默认为代理模式），由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5143 中完成\n* MCP 工具管理器，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5230 中完成\n* 智能技能选择，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5236 中完成\n* README 更新，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5238 中完成\n* 修复（收集器）：对于未明确指定扩展名的 URL，从 Content-Type 推断文件扩展名","2026-04-02T20:56:02",{"id":181,"version":182,"summary_zh":183,"released_at":184},281028,"v1.11.2","## 更多 UI 改进\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F9f2a0363-e905-420e-9c80-3b96fdb07368\n\n现在，在主聊天界面中，我们添加了一些备受期待的 UI 改进和修复。\n- [x] 新的提示输入框\n- [x] 更好的引用 UI 和报告功能\n- [x] 代理调用的指标\n- [x] 在代理调用过程中报告文档和网络搜索引用！\n- [x] 可以在提示中分别开启或关闭代理技能\n- [x] 无需离开页面即可为工作区选择提供商和模型。\n\n----\n\n## 变更内容\n* 5112 或流式指标及完成原因，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5117 中实现\n* 修复了当已存在任何 .env 文件时，`yarn setup:envs` 命令会失败的 bug。由 @brianpursley 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5116 中修复\n* 修复：当 LMStudio 模型列表加载失败或返回空时，显示可操作的错误信息。由 @elevatingcreativity 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5131 中修复\n* 添加自动聊天模式，并支持原生工具调用。由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5140 中实现\n* 侧边栏更新。由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5154 中实现\n* 移除 Google 网络搜索 Programmable SERP。由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5156 中移除\n* 重构：将代理技能设置页面重构为使用 i18n 翻译键。由 @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5146 中实现\n* 杂项：为 `\u002Fcollector` 添加 ESLint。由 @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5128 中完成\n* 杂项：为 `\u002Fserver` 添加 ESLint。由 @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5126 中完成\n* 修复：Azure OpenAI 模型密钥冲突问题。由 @RALaBarge 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5092 中修复\n* 功能：为 paperclip 附件按钮添加提示，当未解析任何文件时显示提示信息。由 @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5139 中实现\n* 修复：为 Confluence Cloud 引用 URL 添加缺失的 `\u002Fwiki` 部分。由 @MaxwellCalkin 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5167 中修复\n* 从复制消息输出中移除“思考”字样。由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5179 中实现\n* 为 Ollama 的嵌入器添加自定义 fetch 实现。由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5180 中实现\n* 杂项：添加脚本以检测并清理未使用的翻译键。由 @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5141 中完成\n* 杂项：添加 ESLint CI 工作流。由 @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5160 中完成\n* 实现 v2 聊天布局设计。由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5074 中实现\n* 改进 zh_TW 正体中文本地化。由 @PeterDaveHello 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5187 中改进\n* 改善测试和 lint 的构建时间。由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F5193 中改进\n* 支持 Agent s","2026-03-18T16:57:50",{"id":186,"version":187,"summary_zh":188,"released_at":189},281029,"v1.11.1","## 首页重新设计\n\nAnythingLLM 的主页面已进行全面重新设计，更加现代化且用户友好，让您在完成初始设置后打开应用的瞬间即可立即开始聊天。\n\n\u003Cimg width=\"3680\" height=\"2212\" alt=\"homepage\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fb7b202b1-3a22-42c9-b46e-434b694c2ec5\" \u002F>\n\n\n## 原生工具调用\n\n\u003CCallout type=\"info\">\n  原生工具调用是利用您的 LLM 提供商和模型进行工具调用时性能与体验的最佳方式。如果可以启用，您应该启用它。\n  \n  _这仅适用于本地 LLM 提供商。对 OpenAI、Anthropic 或 Azure 等云端 LLM 没有任何影响。_\n\u003C\u002FCallout>\n\n我们已彻底重构了 `@agent` 工具调用的工作方式。现在，我们将充分利用您的 LLM 提供商和模型所具备的原生工具调用能力。\n\n**这对您意味着：**\n- 您现在可以使用您的 LLM 提供商和模型执行复杂的 **多步骤** 工具调用。\n- 您的模型将继续运行，直到生成最终响应或被判定为已完成。\n- 即使是小型工具调用模型，也能提供 100 倍更优质的响应。\n\n我们还实施了安全措施，以防止无限循环，每条响应最多允许 10 次工具调用，从而避免任务失控。\n\n\u003Cimg width=\"3680\" height=\"2212\" alt=\"native-tool-calling\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F52676899-77ca-4fdc-b728-9d61c7834536\" \u002F>\n\n\n### 局限性\n\n大多数提供商不允许我们检测模型是否支持原生工具调用。\n\n以下本地 LLM 提供商将在您的模型支持的情况下自动启用原生工具调用：\n- 默认内置 LLM 提供商（AnythingLLM 默认）\n- Ollama\n- LM Studio\n\n对于其他提供商，您需要设置一个环境变量来为支持的提供商启用原生工具调用：\n- Generic OpenAI\n- Groq\n- AWS Bedrock\n- Lemonade\n- LiteLLM\n- Local AI\n- OpenRouter\n\n这可以通过 [`PROVIDER_SUPPORTS_NATIVE_TOOL_CALLING`](\u002Fconfiguration#native-tool-calling-for-llm-providers) 环境变量进行设置。\n``` \nPROVIDER_SUPPORTS_NATIVE_TOOL_CALLING=\"bedrock,generic-openai,groq,lemonade,litellm,local-ai,openrouter\"\n```\n\n## AMD 的 Lemonade 集成\n\u003Cimg width=\"3440\" height=\"1624\" alt=\"lemonade\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F9f1b56bc-0ad4-48ac-972d-a6541b2225f7\" \u002F>\n\nAMD 的 [Lemonade](https:\u002F\u002Flemonade-server.ai) 是一个 [开源](https:\u002F\u002Fgithub.com\u002Flemonade-sdk\u002Flemonade) 的本地模型运行时，专为本地模型（LLM、ASR、TTS、图像生成等）优化性能和效率，适用于包括 AMD GPU 和 NPU 在内的各种硬件。\n\n我们现已添加了一流的支持，使您能够通过 Lemonade 运行的本地模型在 AnythingLLM 中使用，从而在您的本地硬件上获得最佳的应用体验。\n\n\n-----\n\n## 变更内容\n* 修复：贡献指南中的拼写错误，更新项目元数据及 pull_request_temp...md 文件，由 @dipanshurdev 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002F 中完成。","2026-03-02T17:26:08",{"id":191,"version":192,"summary_zh":193,"released_at":194},281030,"v1.11.0","# AnythingLLM 桌面叠加层已上线！\n> _这是一项免费且仅限桌面端的功能！_\n\n现在，AnythingLLM 桌面版配备了一个操作系统级别、具备应用感知的面板，只需一键即可打开。您可以无缝地将当前打开的应用程序与文档聊天、RAG、智能体等其他聊天功能结合使用。\n\n这个面板让使用 AnythingLLM 变得更加流畅便捷——我们强烈推荐您将其用于日常使用！\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F0d3c260a-4d55-46e8-b11a-c6ecbf63e94d\n\n-----\n\n## 变更内容\n* @j0rDy 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4894 中更新了 common.js\n* @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4870 中新增了滚动到聊天记录顶部和底部的快捷键\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4904 中修复了 YouTube 文件收集器 API 的问题\n* @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4828 中优化并标准化了用户名约束条件\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4862 中设计了全新的登录页面 UI\n* [杂项] @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4907 中开发了基于 DMR 的自动翻译工具\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4908 中实现了在 LC 工具中持久化 Ollama 上下文偏好设置\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4909 中重构了 Ollama 上下文窗口设置\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4922 中将 Azure OpenAI 工具调用从 `function` 改为 `tool`\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4916 中实现了思考块状态的持久化切换\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4926 中通过数据库标志管理引导流程的决策\n* @17ColinMiPerry 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4881 中修复了侧边栏线程层级的问题\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4890 中开发了切换开关组件\n* @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4914 中修复了“用户名已存在”的错误\n* @naaa760 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4897 中修复了流式对话期间引用 UI 出现卡顿的问题\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4937 中支持 PrivateModeAI 集成\n* @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4848 中新增了编辑现有 SQL 智能体连接的功能\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4933 中修复了智能体流程构建器中的双层滚动容器问题\n* @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4923 中启用了必要的 ESLint 规则，并重构了前端 lint 配置\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4943 中实现了 SambaNova 集成\n* 更新 LMStudio","2026-02-18T17:08:45",{"id":196,"version":197,"summary_zh":198,"released_at":199},281031,"v1.10.0","\u003Cimg width=\"2560\" height=\"700\" alt=\"v1 10 0 (1)\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F39285372-787a-4b2d-b493-06e9d8044932\" \u002F>\n\n## 重点更新\n\n### AnythingLLM 桌面助手正式上线！\n现在，AnythingLLM 桌面版可以无缝替代 Granola、Otter、Fireflies 等付费工具。\n- 完全在本地设备上运行，无需加入会议即可录制，还能对任意文件进行摘要生成。\n- 基于 NVIDIA Parakeet 技术与 AnythingLLM 的本地编排能力。\n- 可调用任何代理工具、MCP 或您已使用的其他服务，一切通过 AnythingLLM 即可实现！\n- 支持自定义摘要模板、与会议记录对话，甚至具备发言者识别功能。\n- 提供“已加入会议”桌面通知，只需点击一下即可开始新录制。适用于各类会议软件（Zoom、Slack、Discord、Teams 等）。\n- **无速率限制、使用上限或任何限制**\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F355ad2fc-cca1-4e9d-8f08-4a1a08dac48d\n\n\n### AnythingLLM 移动端已在 Google Play 上线\n[Android 版 AnythingLLM 移动应用](https:\u002F\u002Fplay.google.com\u002Fstore\u002Fapps\u002Fdetails?id=com.anythingllm)现已在 Google Play 商店上线。该应用可与 AnythingLLM 的云端\u002F自托管版本及桌面版本实现同步。\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fe4232dfa-e42e-4d15-af9e-d58f2f0245f9\n\n## 其他重要变更\n- 移除新手引导中的“创建工作空间”页面，用户将直接进入主页，并以母语显示新工作空间。\n- 重构工作空间文件选择器，提升性能。\n- 将 Azure OpenAI 迁移到统一的 `v1` API，全面支持代理功能。\n- 修复 paperless-ngx 中的分页错误。\n- 解决因未公开的 YouTube API 变更导致 YT 抓取功能失效的问题。\n- 新增 Cohere 作为代理提供商。\n- 更新多项依赖库版本。\n- 修复拖放 XLSX 文件到聊天窗口后模型无法识别的问题。\n- 修正非 Windows 系统中 MCP 路径的相关问题。\n- 更新 Docker 镜像并打上补丁，使 Scout 评分达到 B 级。\n- 在 UI 中添加错误边界，防止白屏崩溃。\n\n## 具体变更内容\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4746 中将 `body-parser` 补丁版本升级。\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4747 中移除收集器中的 bcrypt，因为未被使用。\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4748 中升级 `jws` 版本。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4740 中重构 WorkspaceFileRow 组件。\n* @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4751 中通过在应用启动时刷新用户数据，修复 UI 中过时的用户权限问题。\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4754 中将 Docker 基础镜像升级至 Ubuntu 24。\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4760 中将 expressJS 从 4.18.2 升级至 4.21.2。\n* 将 Azure OpenAI 集成迁移至 v1 API | 在 Azure OpenAI 基础推理提供程序中启用推理模型的流式传输功能，由 @angelplus 实现。","2026-01-22T15:52:48",{"id":201,"version":202,"summary_zh":203,"released_at":204},281032,"v1.9.1","## 重要变更\n* **PWA 支持**：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4664\n* 在挂载时启用方向键导航斜杠命令：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4543\n* 增加在上传器和聊天中自动处理 YouTube 视频链接的功能：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4547\n* @shatfield4 实现的无纸化 ngx 数据连接器：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4121\n* @alexbarron 添加的 SerpApi 网页搜索功能：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4623\n* @shatfield4 提供的 Z.ai LLM 和代理服务：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4573\n\n----\n\n## 已更改内容\n* @shatfield4 修复 KoboldCPP 代理服务的最大令牌数问题：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4519\n* @shatfield4 启用 Microsoft Foundry Local 作为工作区代理服务：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4539\n* @timothycarambat 实现挂载时使用方向键导航斜杠命令的功能：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4543\n* @timothycarambat 增加上传器和聊天中自动处理 YouTube 视频链接的能力：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4547\n* @aeromech-1 修复服务器端 Qdrant 大规模上传的批处理逻辑：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4545\n* @shatfield4 实现支持变量扩展的代理工作区系统提示：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4526\n* @angelplusultra 重构 `DefaultChatContainer`，显示简单的欢迎消息：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4542\n* @angelplusultra 重构 `loadYouTubeTranscript()`，当 `parseOnly` 为 `true` 时，在内容中包含 YouTube 视频元数据：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4552\n* @shatfield4 修复代理思考 UI 的 bug：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4549\n* @AdeboyeDN 实现对 Northflank 部署的支持：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4570\n* @angelplusultra 在 `\u002Fcollector` 中安装 `patch-package` 并应用补丁以修复 EPub 上传 bug：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4630\n* @timothycarambat 实现无效向量数据库标识符的故障转移至 LanceDB：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4661\n* @naaa760 修复邀请链接用户创建时的无意义错误信息：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4621\n* @naaa760 实现允许使用电子邮件地址作为用户名的功能：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4619\n* @timothycarambat 修复 GitLabLoader 中 `fetchNextPage` 使用的循环逻辑：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4662\n* @timothycarambat 为 Markdown 渲染器添加 Svelte 渲染功能：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4663\n* @angelplusultra 在入口 Shell 脚本中禁用 Prisma CLI 的遥测功能：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4609\n* 支持 PWA 安装","2025-12-09T20:56:11",{"id":206,"version":207,"summary_zh":208,"released_at":209},281033,"v1.9.0","## `@agent` 全面升级与流式传输 ⚡️️\n\n\u003Cvideo src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Ffbbfab37-e1b0-442b-b7d4-277aacfcf9bb\">\u003C\u002Fvideo>\n\n当 AnythingLLM 刚刚推出时，“代理”这个词在大模型领域还并不常见。然而，如今代理正迅速成为构建 AI 应用的标准方式，同时也是与大模型交互的核心体验。\n\n长期以来，由于构建代理的复杂性、工具调用支持不完善、部分模型甚至无法使用工具等技术难题，我们往往只能接受一种不够流畅的使用体验：99% 的时间里，用户都在盯着加载动画，等待响应结果。\n\n### 全新的代理体验现已上线\n\n实时流式传输工具调用和响应（所有服务商、所有模型均可支持）  \n代理现在可以实时从网络下载并解析文件（例如 PDF、Excel、CSV 等）。任何需要处理文档的场景，代理都可以直接从网上实时读取内容。\n\n### 即将推出的功能：\n- 无需代理流程即可实现代理的实时 API 调用  \n- 代理图像理解能力  \n- 代理系统提示符透传 + 用户上下文感知  \n- 跨平台默认技能：实时文件搜索  \n\n## 值得关注的改进：🚀\n * 所有模型和服务商现均已支持代理式流式传输  \n * 集成 Microsoft Foundry Local  \n * 通过代理或上传工具临时抓取\u002F下载任意网络资源  \n\n## 变更内容\n* 允许默认用户重新排序工作空间，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4292 中实现  \n* 为 JSON 和 JSONL 格式添加图片导出支持，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4359 中实现  \n* 修复：提示词缺少编辑图标的问题，由 @17ColinMiPerry 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4344 中修复  \n* 功能 (i18n)：添加缺失的葡萄牙语（巴西）翻译，由 @beckeryuri 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4328 中完成  \n* 功能：实现 CometAPI 集成，用于聊天补全及模型 m…，由 @TensorNull 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4379 中完成  \n* 粘贴时调整聊天文本区域大小，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4369 中实现  \n* 更新保存文件中的代理文本内容，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4389 中完成  \n* 为 document\u002Fupload、document\u002Fupload\u002F{folderName} 和 document\u002Fupload-link 接口新增元数据参数，由 @jstawski 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4342 中添加  \n* 添加对 `SIMPLE_SSO_NO_LOGIN_REDIRECT` 配置项的支持，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4394 中实现  \n* 修复文件夹名称 GET 请求的响应问题，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4395 中完成  \n* 在通用 OpenAI 服务商发送的请求中添加 User-Agent 头信息，由 @angelplusultra 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4393 中完成  \n* 在 API 响应中报告已完成分块的来源信息，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4396 中完成  \n* 允许用户指定","2025-10-09T22:20:12",{"id":211,"version":212,"summary_zh":213,"released_at":214},281034,"v1.8.5","# AnythingLLM v1.8.5 已上线\n\n\n## 重要更新\n\n### 移动端支持\n\n目前，在[实验性功能](https:\u002F\u002Fdocs.anythingllm.com\u002Fbeta-preview\u002Fenable-feature)中，您可以将[AnythingLLM 移动应用 - Android 测试版](https:\u002F\u002Fanythingllm.com\u002Fmobile)连接到您的实例，从而实现设备内与设备外体验的无缝融合。在单一统一的界面中，即可充分利用您实例中的智能体技能和工作流！\n\n\n### 文档聊天功能全面升级\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F0e4ab18d-fb58-480a-94f5-907664dd8f3f\n\n当我们最初构建 AnythingLLM 时，平均上下文窗口仅为 4K，几乎无法容纳整份文档。因此，我们决定始终优先采用 RAG 模式。然而，这种做法存在一些局限性：RAG 的语义依赖性较强，要求用户提出的问题必须与文档内容相关。这导致在尝试“总结这份文档”时，模型往往会回答：“你在说什么？”而文档本身却无法提供进一步的信息。\n\n如今，我们实现了两全其美：文档被限定在特定的工作空间线程和用户范围内，并且我们会尽可能使用完整的文档文本，只要您的模型能够支持。如果超出这一限制，系统会提示您对文档进行嵌入，以便解锁长期记忆功能。\n\n\u003Cimg width=\"1394\" height=\"906\" alt=\"context-warning\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fc28c24d9-a696-4ff6-8d77-7ca708592071\" \u002F>\n\n此外，您还可以轻松管理并查看当前的上下文窗口，移除不再相关的文件，同时保留对话历史记录。\n\u003Cimg width=\"1804\" height=\"732\" alt=\"manage-attached-docs\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F566c6cbd-e67e-4339-82c6-61dfbddde7fd\" \u002F>\n\n\n当然，您仍然可以直接在工作空间的文件管理器中嵌入文件 :)\n\n\n## 其他变更\n* 添加 `model` 字段参数，用于通用 OpenAI TTS 功能，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4171 中实现。\n* 在文本分割器发生更改时清除缓存文档，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4163 中完成。\n* 更新 BARE_METAL.md 文件，由 @buraktopalyn 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4175 中完成。\n* 运行 Jest 测试，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4177 中执行。\n* 新增 Moonshot AI LLM 及智能体提供商，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4178 中实现。\n* 修复通过 OpenAI 兼容 API 实现的多模态聊天问题，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4135 中完成。\n* 原生嵌入模型选择功能（包括多语言支持），由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3835 中实现。\n* 修正翻译字符串及 index.jsx 中的部分拼写和语法错误，由 @antonharder 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4195 中完成。\n* 修改 PostgreSQL SQL 连接器，使其支持查询默认 public schema 以外的其他 schema 中的表，由 @romanwu10 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4202 中完成。\n* 修复应用版本指标端点。","2025-08-15T00:42:56",{"id":216,"version":217,"summary_zh":218,"released_at":219},281035,"v1.8.4","_这是一次小版本更新_\n\n## 重要变更\n\n- 工作区和线程搜索现已移至侧边栏\n- SQL 预检连接验证（终于实现了）\n- 代码块标题在滚动时保持固定\n- 限制代码块的最大宽度，以防止长字符串在界面中溢出\n\n## 具体变更\n* 处理消息响应中的 BigInt 类型，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4110 中完成\n* 为嵌入式聊天组件添加可配置的消息数量限制，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4114 中完成\n* 增加工作区和线程搜索功能，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4120 中完成\n* 文档（i18n）：补全缺失的中文翻译，由 @zz22558822 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4123 中完成\n* 修复分块\u002F片段日志，使其更清晰，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4129 中完成\n* 代码块标题固定，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4130 中完成\n* 文档（zh-cn）：更新并优化简体中文翻译，由 @ChenJiahao1 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4137 中完成\n* 对 YouTube 转录文件路径进行安全 sanitization，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4148 中完成\n* 添加在 MSSQL Server 数据库上启用加密的选项，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4134 中完成\n* 将文本大小菜单迁移到内嵌工具提示，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4147 中完成\n* 设置代码块最大宽度以防止溢出，由 @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4157 中完成\n* SQL 预检连接验证，由 @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4150 中完成\n\n## 新贡献者\n* @zz22558822 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4123 中完成了首次贡献\n* @ChenJiahao1 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4137 中完成了首次贡献\n\n**完整变更日志**：https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fcompare\u002Fv1.8.3...v1.8.4","2025-07-16T18:06:25",{"id":221,"version":222,"summary_zh":223,"released_at":224},281036,"v1.8.3","# AnythingLLM 1.8.3 已上线！\n\n## 新闻\n别忘了报名参加 [AnythingLLM 移动版测试版](https:\u002F\u002Fanythingllm.com\u002Fmobile)！\n- 在设备上运行小型大模型，支持完全本地的 RAG、智能体工具等功能。全程私密，所有处理均在设备端完成。\n- 支持通过本地网络实现移动端与桌面端之间的设备间同步。\n\n## 桌面端重要更新\n- [**内置浏览器工具**](https:\u002F\u002Fdocs.anythingllm.com\u002Ffeatures\u002Fbrowser-tool) 现在你可以抓取私有、需认证或其他受限制的内容，用于智能体的网页抓取或工作流中！\n- [完整变更日志请见此处](https:\u002F\u002Fdocs.anythingllm.com\u002Fchangelog\u002Fv1.8.3)\n\n## 重要更新与修复\n- 轻松将智能体工作流、Slash 命令和系统提示词发布到 [AnythingLLM 社区中心](https:\u002F\u002Fhub.anythingllm.com)。\n- 修复了因 Google API 变更而失效的 YouTube 字幕提取工具。\n\n## 具体变更\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3984 中添加了用于禁用登录界面及基于凭据认证端点的标志。\n* @kilavvy 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3987 中修正了注释和文档中的错别字。\n* @vtjl10 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3993 中修正了 MetaGenerator.js 文档注释中的错别字。\n* @zeevick10 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4003 中修复了 Milvus 提供程序中的错别字，并更新了 Prisma Schema 中的注释。\n* @maximevtush 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4009 中修复了博客文章示例以及 save-file-browser 插件中的错别字。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3976 中实现了将系统提示词发布到社区中心的功能。\n* @leopardracer 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4015 中修正了注释和文档中的错别字。\n* @timothycarambat 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4044 中增加了在系统设置侧边栏中显示应用版本的功能。\n* @AlefAlfa 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4029 中补充了缺失的德语翻译。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4036 中修复了嵌入式聊天组件中提示词、模型和温度覆盖的问题。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4001 中实现了将智能体工作流发布到社区中心的功能。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4019 中实现了将 Slash 命令发布到社区中心的功能。\n* @Tarkva 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4067 中添加了爱沙尼亚语翻译。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4078 中将显示滚动条的选项移至聊天设置中。\n* @martin0258 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4074 中更新了 common.js 中针对 zh_TW 的拼写检查。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4079 中添加了通过环境变量禁用流式输出的选项，适用于通用 OpenAI 提供程序。\n* @shatfield4 在 https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F4084 中修复了 LMStudio 中值嵌入的相关问题。\n* 允许静态定义","2025-07-09T19:23:35",{"id":226,"version":227,"summary_zh":228,"released_at":229},281037,"v1.8.2","# AnythingLLM v1.8.2 is LIVE!\r\n\r\n## Other news\r\n\r\nWe were featured on stage at Microsoft Build (how cool???)\r\n\r\n![1747677767492](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F1e77b3aa-8232-4050-b471-d2eed68ee3f6)\r\n\r\n## Notable Changes\r\n\r\n### Model swap in chat\r\nYou can now easily swap model or provider in the middle of chatting via `cmd\u002FCtrl + Shift + L` to show a tooltip menu to easily swap models without changing screens\r\n\r\n\u003Cimg width=\"1132\" alt=\"Screenshot 2025-06-10 at 4 56 29 PM\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F400443ff-4cda-4605-a85b-ac24b80b1e63\" \u002F>\r\n\r\n\r\n### System Prompt version tracking\r\nNow, when you edit the system prompt of any workspace we locally store this information so you can easily refer to or restore to previous prompts that might work better for a given model or workspace.\r\n\r\n\u003Cimg width=\"1138\" alt=\"Screenshot 2025-06-10 at 4 56 21 PM\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F6c121735-8e61-4dff-a1e2-88e659071b10\" \u002F>\r\n\r\n## What's Changed\r\n* Fix\u002Fdom nesting warnings by @17ColinMiPerry in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3773\r\n* Refactor AWS Bedrock Provider for Multi-modal Support & Correct Token Limits by @tristan-stahnke in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3714\r\n* feat: support for iam roles for bedrock client by @dannysteenman in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2632\r\n* feat: support fetch wikis for gitlab data connectors by @lcgash in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3271\r\n* Updated README.md and README.zh_CN.md by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3775\r\n* Updated README.md and README.zh-CN.md by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3776\r\n* Enable workflow rule for package verification by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3778\r\n* Update README.zh-CN.md by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3780\r\n* Chat prompt history menu by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3770\r\n* minor UI patches by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3784\r\n* Updated zh locale by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3786\r\n* locale change by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3787\r\n* Add Disable Spellcheck Option by @17ColinMiPerry in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3795\r\n* Update:WorkspaceChat-ChatContainer-PromptInput by @maoguy in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3792\r\n* updated locale by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3796\r\n* PGvector vector database support by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3788\r\n* Fix MCP Start\u002FStop Toast Text by @17ColinMiPerry in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3803\r\n* Fixed grammar and typos by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3802\r\n* Obsidian data connector by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3798\r\n* Sponsors by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3816\r\n* Fix typos by @omahs in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3826\r\n* Update light mode designs for system prompt history by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3818\r\n* Warn bad docker command by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3827\r\n* Automatic Context window detection by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3817\r\n* Improve UX for API keys and invitations by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3828\r\n* Support Dell Pro AI Studio provider by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3829\r\n* Optional summarization for web-scraping flow block by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3496\r\n* 3416 file picker load times by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3833\r\n* Respect Max Chunk Len set - Generic OpenAI Embedder connection by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3838\r\n* Disable message send while content is embedding by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3832\r\n* Updated Chinese locale by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3852\r\n* Filter malformed post-processed docs from file picker by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3853\r\n* Update Brazilian Portuguese locale (pt_BR) by @JairoMarques in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3879\r\n* Removed redundant scrollbar in search containers by @17ColinMiPerry in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3865\r\n* Fix File attachments are not cleared on second send click when use @a… by @cnjsyyb in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3843\r\n* Update cit","2025-06-10T23:59:29",{"id":231,"version":232,"summary_zh":233,"released_at":234},281038,"v1.8.1","## What's Changed\r\n* enable MCP tool use via `@agent` in API calls by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3683\r\n* Add support for `\u002Fv1\u002Fdocument\u002Fupload*` endpoints to support auto-add to workspace by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3692\r\n* Feature\u002Fdrupalwiki collector by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3693\r\n* Enable bypass of ip limitations via ENV in collector processing by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3652\r\n* Translate main page landing by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3699\r\n* Allow custom headers in upload-link endpoint by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3695\r\n* Improve Japanese translation to maintain @agent by @suzuki-shm in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3704\r\n* Update `embed` by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3728\r\n* TTS auto-speak and STT auto-submit by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3696\r\n* Update Bing Search API Azure Service copy - Unavailable by @thejjw in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3729\r\n* Update light mode input color by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3734\r\n* Replace @azure\u002Fopenai with openai, and update openai to version 4.95.1 by @suzuki-shm in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3691\r\n* Remove `@google\u002Fgenerative-ai` from AnythingLLM by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3740\r\n* Update README.zh-CN.md by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3748\r\n* pt-BR translation by @thisisleobro in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3755\r\n* #3707 added sse and streamable stransport support for mcp servers by @AlekseyKapustyanenko in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3747\r\n* App wide antialiasing by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3752\r\n* Add option to control KoboldCPP max response tokens by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3746\r\n* handle no thoughts in container by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3758\r\n* Remove drop shadow from CTAButton component by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3756\r\n* Fix chat alignment borders by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3760\r\n* Data connector CTA button styles by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3759\r\n* Add gemini-embedding-exp-03-07 model support by @sdarioz in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3767\r\n* Filter image models from OpenAI by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3768\r\n* more zh translations by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3769\r\n\r\n## New Contributors\r\n* @thejjw made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3729\r\n* @thisisleobro made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3755\r\n* @AlekseyKapustyanenko made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3747\r\n* @sdarioz made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3767\r\n\r\n## Desktop App Changelog\r\nhttps:\u002F\u002Fdocs.anythingllm.com\u002Fchangelog\u002Fv1.8.1\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fcompare\u002Fv1.8.0...v1.8.1","2025-05-06T17:50:20",{"id":236,"version":237,"summary_zh":238,"released_at":239},281039,"v1.8.0","# AnythingLLM v1.8.0 | MCP tools & a fresh new look!\r\n\r\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F5639b994-2739-4605-9eee-a923148ceb3a\r\n\r\n## MCP tooling is now easier than ever!\r\n\r\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F4a62acc7-ec34-4761-b662-29aa48194db1\r\n\r\n- [Full Documentation](https:\u002F\u002Fdocs.anythingllm.com\u002Fmcp-compatibility\u002Foverview)\r\n- Update and reload your MCP config on the fly with no downtime\r\n- Docker includes `npx` and `uv` commands build in\r\n\r\n## A new onboarding experience\r\n\r\n\u003Cimg width=\"1612\" alt=\"Screenshot 2025-04-18 at 3 16 46 PM\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fe690a80f-9b41-4443-a739-5be35d312fcf\" \u002F>\r\n\r\nAin't it pretty?\r\n\r\n\r\n## What's Changed\r\n* Replace connectionString parser for sql-agent by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3560\r\n* Update styles of experimental modal buttons by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3535\r\n* Teal color for onboarding buttons by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3533\r\n* Agent builder UI bug fixes by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3522\r\n* ModelContextProtocol (MCP) Full Compatibility by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3547\r\n* Fix broken links in MCP UI by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3569\r\n* Allow period in username by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3592\r\n* Fix agent flow description bug by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3593\r\n* Sync and publish `embed` widget changes for thought models by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3596\r\n* [CHORE] Consistent table styling across settings pages by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3597\r\n* Gemini model list sync by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3609\r\n* Disable Agent retry to prevent looping on crash by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3617\r\n* Refactor Gemini to use OpenAI interface API by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3616\r\n* Fix bottom border on workspace settings tab switcher by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3615\r\n* Move CTA save buttons to top of workspace setting screens by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3613\r\n* Fix truncation of threads\u002Fworkspaces by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3610\r\n* New chat home page by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3555\r\n* Main page QA by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3623\r\n* Appearance menu styles by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3627\r\n* Fix customize logo bug by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3628\r\n* Add Japanese translations for AnythingLLM by @suzuki-shm in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3651\r\n* handling of citations in openRouter provider #3581 by @rudimichal in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3620\r\n* Buildx bump to latest to solve GHA cache deprecation by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3653\r\n* Apply `.git` logic handler for repo URLs by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3655\r\n* Updated Russian translation by @aguilam in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3658\r\n* Update Azure AI options and model map with new model configurations by @suzuki-shm in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3660\r\n\r\n## New Contributors\r\n* @suzuki-shm made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3651\r\n* @rudimichal made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3620\r\n* @aguilam made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3658\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fcompare\u002Fv1.7.8...v1.8.0","2025-04-18T22:17:11",{"id":241,"version":242,"summary_zh":243,"released_at":244},281040,"v1.7.8","## What's New\r\n\r\n- Support for Workspace [system prompt variables!](https:\u002F\u002Fdocs.anythingllm.com\u002Ffeatures\u002Fsystem-prompt-variables)\r\n- NVIDIA NIM support - [AnythingLLM Desktop only](https:\u002F\u002Fdocs.anythingllm.com\u002Fnvidia-nims\u002Fintroduction)\r\n\r\n_this version includes 1.7.7, which no changelog was published for - whoops!_\r\n\r\n## What's Changed\r\n* Enable slash commands in dev API chat\u002Fstream-chat by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3514\r\n* Enable use of @agent in slash commands by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3508\r\n* Protected slash command check by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3524\r\n* Prompt variables by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3359\r\n* Update QDRANT_SETUP.md by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3530\r\n* Overhaul AWS Bedrock provider for `AIProviders` by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3537\r\n* add AgentFlow execution via API chat invocation by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3545\r\n* Persist images attached in API requests to the UI for rendering by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3546\r\n\r\n_1.7.7 changes_\r\n* CHORE: debounce telems to prevent spamming by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3505\r\n* Bump LanceDB by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3015\r\n* Update README.zh-CN.md by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3527\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fcompare\u002Fv1.7.6...v1.7.8","2025-03-27T19:59:37",{"id":246,"version":247,"summary_zh":248,"released_at":249},281041,"v1.7.6","## What's Changed\r\n* fix thought container from getting cut off at max height by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3449\r\n* Fix agent calling UI max height by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3457\r\n* Added danish translation by @MOEI-123412 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3461\r\n* Pinning UI improvements by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3490\r\n* Add remove folder developer API endpoint by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3489\r\n* fix chroma db + add similarity offset by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3458\r\n* Fix embedding endpoint for OpenAI compatible API by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3467\r\n* Add ability to reordering workspaces by @MrMarans in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3446\r\n* Use concurrently to run all dev service at once by @dotku in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3424\r\n\r\n## New Contributors\r\n* @MOEI-123412 made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3461\r\n* @dotku made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3424\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fcompare\u002Fv1.7.5...v1.7.6","2025-03-19T22:52:50",{"id":251,"version":252,"summary_zh":253,"released_at":254},281042,"v1.7.5","## Notable Changes\r\n\r\n- Anthropic provider now uses their `\u002Fmodels` endpoint (finally)\r\n- Bug fixes on collector to prevent thread crashing\r\n- Multi-language OCR support on collector\r\n- More localization and language support\r\n\r\n## What's Changed\r\n* Add additional embed API endpoints by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3264\r\n* feat: Add endpoint to retrieve documents by folder name by @thestackdev in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3258\r\n* Improve PromptInput component by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3266\r\n* 2749 ollama client auth token by @ssbodapati in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3005\r\n* Normalize chat sorting defaults in developer API by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3270\r\n* Markdown support in custom messages by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3267\r\n* Update null-check function for aibitat runners by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3283\r\n* Fix garbled non English chars on document upload by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3301\r\n* [FEAT] Add claude-3-7 by @kaashmonee in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3337\r\n* feature: support configurable left and right message layout by @MrGaoGang in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3244\r\n* Arabic Translation by @Saidoua in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3357\r\n* fix: bug where in single user mode deletion of an api key deletes all  by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3358\r\n* feat: Add endpoint to upload documents to a specified folder by @thestackdev in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3276\r\n* Add bio field to user by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3346\r\n* speech to text typo correction by @yashschandra in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3330\r\n* New thinking\u002Fagent animation + UI by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3302\r\n* Windows development environment variables support by @kbourro in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3354\r\n* Add new model provider PPIO by @cnJasonZ in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3211\r\n* changed language support for following windows: chat, data-upload and… by @MrMarans in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3010\r\n* feat: Add multilingual support for ocr module by @doodle777 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3325\r\n* Prevent collector crash when blocked by CDN by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3373\r\n* Add support for Anthropics \u002Fmodel endpoint (finally) by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3376\r\n\r\n## New Contributors\r\n* @thestackdev made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3258\r\n* @ssbodapati made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3005\r\n* @kaashmonee made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3337\r\n* @MrGaoGang made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3244\r\n* @Saidoua made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3357\r\n* @yashschandra made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3330\r\n* @kbourro made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3354\r\n* @cnJasonZ made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3211\r\n* @MrMarans made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3010\r\n* @doodle777 made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3325\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fcompare\u002Fv1.7.4...v1.7.5","2025-03-04T22:14:53",{"id":256,"version":257,"summary_zh":258,"released_at":259},281043,"v1.7.4","## The no-code agent skill builder is live ⚡ \r\n![AnythingLLM](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F14b09cc2-4932-47c1-a81d-5b96ab64504a)\r\n\r\n## Big changes\r\n\r\n### 🏗️ The built-in no-code agent flow builder is alive!\r\n\r\nNow, without a line of code you can seamlessly draft and run agentic skills that _just make sense_. No confusing setup - just AI skills on steroids, fully local and about 1000x easier to understand than MCP.\r\n\r\n[Read the docs on making your first skill ->](https:\u002F\u002Fdocs.anythingllm.com\u002Fagent-flows\u002Ftutorial-hackernews)\r\n\r\n### OCR scanning of PDF and images is live!\r\n\r\nNow, when Anythingllm is given an image or _nondigitized_ PDF we will automatically OCR the text content into a document. This is done via a local [Tesseract](https:\u002F\u002Fgithub.com\u002Ftesseract-ocr\u002Ftesseract?tab=readme-ov-file#installing-tesseract) model running in process!\r\n\r\n_this is OCR, not \"vision\" so if the content has no text - nothing will be found!_\r\n\r\n\r\nLastly, we skipped a few Docker versions to make Docker \u003C=> Desktop the same release version. It was confusing have two different semantic versions. They will now be lock-step going forward.\r\n\r\n## All the other fixes 🔧 \r\n* Enable `include_reasoning` in request by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3067\r\n* parse streams from `reasoning_content` from DeepSeek API by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3068\r\n* Add tokenizer improvements via Singleton class and estimation by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3072\r\n* autodetect parseable text file contents by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3079\r\n* inherit font size from UI for thoughts by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3099\r\n* Enable consistent styling on chart items by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3126\r\n* Add reasoning flag for azure models with \"default\" fallback by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3128\r\n* Enable `num_ctx` to match defined chunk length in ollama embedder by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3129\r\n* Patch PPLX streaming for timeouts by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3130\r\n* Add improved logging and introspection to web-browsing by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3140\r\n* [FEAT] Add Turkish Language Support (tr.js) #3108 by @TunahanDilercan in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3141\r\n* [CHORE] Onboarding language translations by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3176\r\n* fix UserMenu rendered twice on Main page by @sdeleeuw in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3110\r\n* Add handling to reasoning models for Generic OpenAI connector by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3183\r\n* Agent flow builder by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3077\r\n* chore: rename Github to GitHub by @setchy in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3199\r\n* chore: rename `Gitlab` to `GitLab` by @setchy in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3200\r\n* Add `querySelectorAll` capability to web-scraping block by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3186\r\n* Turkish translations by @cobanov in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3207\r\n* OCR PDFs as fallback during upload by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3204\r\n* Add OCR of image support by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3219\r\n* feat(dataconnectors): support confluence personal access token by @lcgash in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3206\r\n* fix agent builder move block bug by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3220\r\n* Drop Node `canvas` for manual `sharp` conversion by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3221\r\n* for projects in gitlab subgroup (#3075) by @t2tx in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3247\r\n* Fix: fix text and format issues in Chinese documentation. by @alongLFB in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3249\r\n* fix: sanitizeNamespace by @hakeemsyd in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3246\r\n* fix: Patch agent flow to honor agent handler established provider by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3251\r\n\r\n## New Contributors\r\n* @TunahanDilercan made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3141\r\n* @sdeleeuw made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3110\r\n* @setchy made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F3199\r\n* @cobanov made their first cont","2025-02-18T01:15:43",{"id":261,"version":262,"summary_zh":263,"released_at":264},281044,"v1.4.0","# 🧠 AnythingLLM is `\u003Cthink>`ing....\r\n\r\n> [!Note]\r\n> **Desktop build of this release (1.7.3) are attached as binaries below**\r\n\r\n## Notable changes 🌟 \r\n- Better Agent calling UI\r\n\u003Cimg width=\"1612\" alt=\"agent-ui\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fac33824f-9e8e-4292-ac3c-9e48630cdcb0\" \u002F>\r\n\r\n- UI support for `\u003Cthink>`ing models in chat response\r\n\u003Cimg width=\"1612\" alt=\"think\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F0af37a48-0bf2-4da6-8980-5bba80afe555\" \u002F>\r\n\r\n## Misc Fixes and improvements 🔧 \r\n\r\n* Add Support for NVIDIA NIM by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2766\r\n* update chat model filters for openai by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2803\r\n* Fixed a typo in README by @logscore in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2805\r\n* Fix scraping failed bug in link\u002Fbulk link scrapers by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2807\r\n* Fix timestamp bug in embed widget by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2806\r\n* Redirect to default thread on thread deletion by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2808\r\n* Add support for hiding sidebar by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2809\r\n* Add vector search API endpoint by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2815\r\n* patch ZH typo by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2816\r\n* add Gitlab to watchable documents by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2817\r\n* Fix(2753): Replace React Router Navigate with window.location for SSO state refresh by @andydev404 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2818\r\n* Docker Scout CI error by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2820\r\n* APIPie LLM provider improvements by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2695\r\n* Add thread name to workspaces get endpoint by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2822\r\n* Add new workspace user management endpoint by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2842\r\n* Add attachments to GenericOpenAI prompt by @wolfganghuse in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2831\r\n* Purge cached docs and remove docs from all workspaces on vectorDB\u002Fembedder changes by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2819\r\n* LLM performance metric tracking by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2825\r\n* Append writable fields to dev API new workspace endpoint by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2843\r\n* Enable `default` roles to be able to use `attachments` but not upload documents by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2860\r\n* Add support for gemini authenticated models endpoint by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2868\r\n* Remove fine-tuning flow by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2872\r\n* fix wrong metadata assignment in MilvusProvider by @wolfganghuse in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2870\r\n* update ollama performance mode by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2874\r\n* Fix light mode table colors by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2873\r\n* Fix styles of ChangeWarningModal by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2876\r\n* Show scrollbar on rendered tables in chat by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2884\r\n* Fix typo in frontend for SearXNG agent setup by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2917\r\n* Normalize paths on files uploaded to prevent arbitrary file writes by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2905\r\n* Patch unauthorized access to other user's pfps by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2904\r\n* Audio file validations by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2902\r\n* Patch custom models endpoint by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2903\r\n* Add support for Google Generative AI (Gemini) embedder by @007900 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2895\r\n* Return timestamps in OpenAI compat endpoints in proper format by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2921\r\n* Update README.md by @KPCOFGS in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2924\r\n* make `similarityResponse` object arguments and not positional by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2930\r\n* Reranker option for RAG by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2929\r\n* Update example ENV fil","2025-01-29T23:40:58",{"id":266,"version":267,"summary_zh":268,"released_at":269},281045,"v1.3.0","# The AnythingLLM Community hub is live!\r\n\r\n\u003Cimg width=\"1204\" alt=\"Share Image\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F78e3b5b6-7611-464e-9ef8-be497cac697a\">\r\n\r\nThe [AnythingLLM Community Hub](https:\u002F\u002Fhub.anythingllm.com) is a platform that works _inside_ of AnythingLLM to easy share and import community create system prompts, slash commands, and **agent skills**.  Everything on the community hub can be shared publicly with everyone or **only privately** with your team.\r\n\r\nIn the near future, we will be enabling:\r\n- Workspace sharing\r\n- Data Connectors\r\n- Authentication flows\r\n\r\n[Learn more about the community hub](https:\u002F\u002Fdocs.anythingllm.com\u002Fcommunity-hub\u002Fabout) and how to pull in or upload your own content to share with others.\r\n\r\n## Other notable changes\r\n\r\nAnythingLLM now has an official Dark and light mode. This is going to pave the way for more custom theme options in the future.\r\n![ezgif com-animated-gif-maker](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F8f940c5e-a95c-4d87-a5e3-a88f596e46cb)\r\n\r\n\r\n## All Changes\r\n* Patch `v1\u002Fdocument\u002Fupload` filename charset encoding by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2535\r\n* Simple SSO feature for login flows from external services by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2553\r\n* add `sessionToken` validation connection auth for AWSbedrock by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2554\r\n* Feature\u002Fupdate api example outputs by @MrSimonC in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2547\r\n* Add optional limit and orderBy to dev api chat endpoint by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2559\r\n* Allow 127.0.0.1 as valid URL for scraping by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2560\r\n* 2488 novita ai llm integration by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2582\r\n* Add header static class for metadata assembly by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2567\r\n* DuckDuckGo web search agent skill support by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2584\r\n* Added translation dutch (nl) by @MelroyB in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2595\r\n* Update GitLab PAT token url by @GabLeRoux in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2598\r\n* Fix the problem of web-searching can not complete searching sessions when using bing search. by @fiyen in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2611\r\n* Handle OpenAI whisper transcription edge case by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2621\r\n* Fix chat window scrollbar appearance option by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2589\r\n* Add undo\u002Fredo functionality #2591 by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2623\r\n* Frontend performance improvements by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2627\r\n* Patch bad models endpoint path in LM Studio embedding engine by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2628\r\n* Bump TogetherAI models by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2645\r\n* Light\u002Fdark mode UI overhaul by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2629\r\n* Pr 2660 branch by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2662\r\n* Add more expiermental models from Gemini by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2663\r\n* Add plaintext file extensions by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2664\r\n* TogetherAI Llama 3.2 vision models support by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2666\r\n* Mistral embedding engine support by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2667\r\n* Pr 2653 branch by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2672\r\n* feat: support setting maxConcurrentChunks for Generic OpenAI embedder by @hdelossantos in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2655\r\n* Adjustable font size in chat input by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2692\r\n* ✨ feat: Add Vietnamese translations and configuration by @bibo318 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2641\r\n* Fix typo with GCP deploy guideline by @TuanBC in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2718\r\n* Fix markdown issue where type is missing in backticks by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2719\r\n* Normalize Input Fields by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2720\r\n* Community hub integration by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2555\r\n* Community hub integration UX improvements by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2727\r\n\r\n## N","2024-11-26T22:44:56",{"id":271,"version":272,"summary_zh":273,"released_at":274},281046,"v1.2.4","## What's Changed\r\n* hide config for agent skills to manager (perm not allowed) by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2469\r\n* Integrate Apipie support directly by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2470\r\n* LiteLLM agent support by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2460\r\n* Pasting text bug fix by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2425\r\n* Fix file name\u002Ffolder name truncation by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2456\r\n* Daily message limit per user by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2417\r\n* Add backfilling on `query` for chat widget to improve UX by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2482\r\n* Support free-form input for workspace model for providers with no `\u002Fmodels` endpoint by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2397\r\n* OpenAI o1 model support by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2427\r\n* Tts open ai compatible endpoints by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2487\r\n* Handle Bedrock models that cannot use `system` prompts by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2489\r\n* Handle non-ascii characters in single and bulk link scraper URLs by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2495\r\n* Passthrough agentModel for LMStudio by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2499\r\n* Patch `DNDFilePicker` useEffect callback by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2506\r\n* Add Groq vision preview support by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2511\r\n* Feature\u002Fthread creation slug name by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2512\r\n* Enable the ability to disable the chat history UI by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2501\r\n* Fix tooling examples by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2513\r\n* Rework failover agent invocation selection for agent and ephemmeral agent by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2515\r\n* Github data connector improvements by @shatfield4 in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2439\r\n* Add Grok\u002FXAI support for LLM & agents by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2517\r\n* Alignment crime fixed by @James-Lu-none in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2528\r\n* Add filtering to sessionID for workspace chats by @timothycarambat in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2531\r\n\r\n## New Contributors\r\n* @James-Lu-none made their first contribution in https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fpull\u002F2528\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FMintplex-Labs\u002Fanything-llm\u002Fcompare\u002Fv1.2.3...v1.2.4\r\n\r\nPinned Desktop version: [v1.6.8](https:\u002F\u002Fdocs.anythingllm.com\u002Fchangelog\u002Fv1.6.8)","2024-10-25T00:02:57"]