[{"data":1,"prerenderedAt":278},["ShallowReactive",2],{"repo-run-llama-llama_index-zh-CN":3,"i18n-run-llama-llama_index-zh-CN":275},{"mirrorPool":4,"detail":13,"relatedProjects":139},[5,9],{"key":6,"label":7,"base":8,"type":6},"github","mirror.github","https:\u002F\u002Fgithub.com\u002F",{"key":10,"label":11,"base":12,"type":10},"auto-proxy","mirror.autoProxy","https:\u002F\u002Fgithub.xxlab.tech\u002F",{"owner":14,"repo":15,"repoId":16,"repoUrl":17,"starCount":18,"language":19,"description":20,"languages":21,"screenshots":28,"homepage":29,"pushedAt":30,"ownerAvatar":31,"ownerName":32,"ownerBio":33,"lastSyncTime":34,"lastReleaseTime":35,"tags":36,"license":47,"category":48,"subgroup":49,"releases":50},"run-llama","llama_index",560704231,"https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index",52579,"Python","LlamaIndex is the document processing platform for AI",{"Python":22,"Jupyter Notebook":23,"Makefile":24,"JavaScript":25,"HTML":26,"Shell":27,"EdgeQL":27,"Starlark":27},72.6,24.9,1.8,0.4,0.3,0,[],"https:\u002F\u002Fdevelopers.llamaindex.ai","2026-10-20","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F130722866?v=4","LlamaIndex","","2026-10-20T10:23:37.432228+00:00","2026-09-21T16:22:27Z",[37,38,39,40,41,42,43,44,45,46],"agents","application","data","fine-tuning","framework","llamaindex","llm","multi-agents","rag","vector-database","MIT","development","开发库\u002F框架",[51,64,75,86,97,108,119,129],{"tag":52,"publishAt":35,"semVer":53,"prerelease":56,"assets":57},"v0.14.25",[27,54,55],14,25,false,[58],{"fileName":59,"sizeMb":60,"sha256":61,"downloadCount":62,"originUrl":63},"llama_index-0.14.25.tar.gz",0.01,"242269bf0dec5f73733572d10866b34aba86f049549fd97612f554461f549e68",35,"https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.25\u002Fllama_index-0.14.25.tar.gz",{"tag":65,"publishAt":66,"semVer":67,"prerelease":56,"assets":69},"v0.14.24","2026-08-19T18:48:01Z",[27,54,68],24,[70],{"fileName":71,"sizeMb":60,"sha256":72,"downloadCount":73,"originUrl":74},"llama_index-0.14.24.tar.gz","9574d3dbd6ba247bd523e78771647807794cce9a47410dbc70ed337bb610a6ff",124,"https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.24\u002Fllama_index-0.14.24.tar.gz",{"tag":76,"publishAt":77,"semVer":78,"prerelease":56,"assets":80},"v0.14.23","2026-06-24T19:36:43Z",[27,54,79],23,[81],{"fileName":82,"sizeMb":60,"sha256":83,"downloadCount":84,"originUrl":85},"llama_index-0.14.23.tar.gz","eac2049816a7410ff4568490cce4bdff99cda3ab99d59f52f6227dad22cda44b",84,"https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.23\u002Fllama_index-0.14.23.tar.gz",{"tag":87,"publishAt":88,"semVer":89,"prerelease":56,"assets":91},"v0.14.22","2026-05-14T20:22:24Z",[27,54,90],22,[92],{"fileName":93,"sizeMb":60,"sha256":94,"downloadCount":95,"originUrl":96},"llama_index-0.14.22.tar.gz","c2c9b31f50d2815abdc191085db4acaf96b7c01851ac66b2e4cc82be8cde589e",89,"https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.22\u002Fllama_index-0.14.22.tar.gz",{"tag":98,"publishAt":99,"semVer":100,"prerelease":56,"assets":102},"v0.14.21","2026-04-21T00:18:51Z",[27,54,101],21,[103],{"fileName":104,"sizeMb":60,"sha256":105,"downloadCount":106,"originUrl":107},"llama_index-0.14.21.tar.gz","99244cbdc7f486aa329c7007faa168085b19eff786ee0c4d246db1cba0f4922b",57,"https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.21\u002Fllama_index-0.14.21.tar.gz",{"tag":109,"publishAt":110,"semVer":111,"prerelease":56,"assets":113},"v0.14.20","2026-04-03T19:55:51Z",[27,54,112],20,[114],{"fileName":115,"sizeMb":60,"sha256":116,"downloadCount":117,"originUrl":118},"llama_index-0.14.20.tar.gz","aa6895cee1366a1ab256715fb2f526d57fe346708c76e77d6f319380de70223b",43,"https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.20\u002Fllama_index-0.14.20.tar.gz",{"tag":120,"publishAt":121,"semVer":122,"prerelease":56,"assets":124},"v0.14.19","2026-03-25T20:59:15Z",[27,54,123],19,[125],{"fileName":126,"sizeMb":60,"sha256":127,"downloadCount":55,"originUrl":128},"llama_index-0.14.19.tar.gz","596e4bf54a4563a9b1eaf41da34b42b861b77691eb08ad7be24165fb9844fee2","https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.19\u002Fllama_index-0.14.19.tar.gz",{"tag":130,"publishAt":131,"semVer":132,"prerelease":56,"assets":134},"v0.14.18","2026-03-16T19:42:07Z",[27,54,133],18,[135],{"fileName":136,"sizeMb":60,"sha256":137,"downloadCount":112,"originUrl":138},"llama_index-0.14.18.tar.gz","43cf535521bec5451aaadfa3dea012f4bf60f7e00db652479332c1acb43d152b","https:\u002F\u002Fgithub.com\u002Frun-llama\u002Fllama_index\u002Freleases\u002Fdownload\u002Fv0.14.18\u002Fllama_index-0.14.18.tar.gz",[140,172,195,223,241,257],{"owner":141,"repo":142,"repoId":143,"category":48,"subgroup":49,"language":19,"tags":144,"starCount":162,"ownerUid":163,"latestTag":164,"platforms":165,"lastReleaseTime":169,"downloadCount":170,"description":171},"unslothai","unsloth",725205304,[145,146,147,148,40,149,150,151,43,152,153,154,155,156,157,158,159,160,161,142],"agent","ai","chatgpt","deepseek","gemma","image-generation","llama","llms","openai","python","qwen","reinforcement-learning","self-hosted","stable-diffusion","text-to-speech","tts","ui",78618,150920049,"v0.1.911-beta",[166,167,168],"windows","mac","linux","2026-10-26T14:23:54Z",12443,"一个用于加速大型语言模型（LLM）微调的开源库，支持大幅降低显存占用并提升训练速度。",{"owner":173,"repo":174,"repoId":175,"category":48,"subgroup":49,"language":19,"tags":176,"starCount":188,"ownerUid":189,"latestTag":190,"platforms":191,"lastReleaseTime":192,"downloadCount":193,"description":194},"openbq-org","OpenBB",323048702,[146,177,178,179,180,181,182,183,184,185,154,186,187],"crypto","derivatives","economics","equity","finance","fixed-income","machine-learning","openbb","options","quantitative-finance","stocks",74466,328144346,"ODP",[166,167],"2026-09-30T01:47:43Z",1099,"OpenBB 是面向分析师、量化研究员和 AI Agent 的开放金融数据平台，以 Python 库与 REST API 形式聚合股票、期权、加密货币、宏观经济等多源数据，一次接入多处消费。",{"owner":196,"repo":197,"repoId":198,"category":48,"subgroup":49,"language":19,"tags":199,"starCount":215,"ownerUid":216,"latestTag":217,"platforms":218,"lastReleaseTime":220,"downloadCount":221,"description":222},"pathwaycom","pathway",571186647,[200,201,202,203,204,205,206,207,208,209,197,154,210,211,212,213,214],"batch-processing","data-analytics","data-pipelines","data-processing","dataflow","etl","etl-framework","iot-analytics","kafka","machine-learning-algorithms","real-time","rust","stream-processing","streaming","time-series-analysis",62092,25750857,"v0.32.1",[219],"other","2026-08-01T09:17:35Z",359,"Pathway 是一个 Python ETL 框架，统一处理批数据与流数据，适用于流处理、实时分析、LLM 管道与 RAG，底层由基于 Differential Dataflow 的 Rust 引擎驱动。",{"owner":224,"repo":225,"repoId":226,"category":48,"subgroup":49,"language":19,"tags":227,"starCount":234,"ownerUid":235,"latestTag":236,"platforms":237,"lastReleaseTime":238,"downloadCount":239,"description":240},"psf","requests",1362490,[228,229,230,231,232,154,233,225],"client","cookies","forhumans","http","humans","python-requests",54522,50630501,"v2.34.2",[219],"2026-05-14T19:27:15Z",1182,"Requests 是 Python 最流行的 HTTP 客户端库，用极简 API 发送 HTTP\u002F1.1 请求，自动处理查询串、表单编码、Cookie、认证与压缩解码。",{"owner":242,"repo":243,"repoId":244,"category":48,"subgroup":49,"language":19,"tags":245,"starCount":250,"ownerUid":251,"latestTag":252,"platforms":253,"lastReleaseTime":254,"downloadCount":255,"description":256},"pandas-dev","pandas",858127,[246,247,248,249,243,154],"alignment","data-analysis","data-science","flexible",50247,21206976,"v3.1.0",[219],"2026-10-22T13:20:42Z",164,"pandas 是 Python 最流行的数据分析与处理库，提供类似 R data.frame 的带标签数据结构 Series 与 DataFrame，覆盖清洗、对齐、分组聚合、时间序列等日常数据工作全流程。",{"owner":258,"repo":259,"repoId":260,"category":48,"subgroup":49,"language":261,"tags":262,"starCount":268,"ownerUid":269,"latestTag":270,"platforms":271,"lastReleaseTime":272,"downloadCount":273,"description":274},"jgthms","bulma",50264296,"CSS",[263,264,265,266,267],"css","css-framework","design","flexbox","html",50039,1254808,"1.0.4",[219],"2025-04-19T14:02:58Z",49173,"Bulma 是一个基于 Flexbox 的现代纯 CSS 框架，无需任何 JavaScript 即可构建响应式界面，提供栅格、组件与修饰符体系，兼容各主流浏览器。",{"description":276,"readmeHtml":277},"LlamaIndex 是一个面向大语言模型应用的开源数据框架（Python），提供数据连接器、索引、查询引擎与 Agent 编排，把私有数据接入 LLM 构建 RAG 与智能体应用。","\u003Ch2 id=\"项目概述\">项目概述\u003C\u002Fh2>\u003Cp>LlamaIndex 是一个面向大语言模型应用的开源数据框架，负责把散落在 PDF、API、SQL、文档等来源的数据接入 LLM。它提供数据连接器、索引、查询引擎与 Agent 编排等一整套工具，是构建 RAG（检索增强生成）与多 Agent 应用的常用底座。框架核心与 300 多个集成包分离发布，可按需挑选偏好的模型与向量库供应商。\u003C\u002Fp>\n\u003Cblockquote>\n\u003Cp>让私有数据成为 LLM 的上下文，一套工具打通接入、索引与问答。\u003C\u002Fp>\n\u003C\u002Fblockquote>\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>项目\u003C\u002Fth>\n\u003Cth>内容\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>平台\u003C\u002Ftd>\n\u003Ctd>Python\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>许可\u003C\u002Ftd>\n\u003Ctd>MIT\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>集成包\u003C\u002Ftd>\n\u003Ctd>300+（LLM、嵌入、向量库等）\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>安装包\u003C\u002Ftd>\n\u003Ctd>llama-index（整合包）\u002F llama-index-core（核心）\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n\u003Ch2 id=\"主要功能\">主要功能\u003C\u002Fh2>\u003Cul>\n\u003Cli>\u003Cstrong>数据连接器\u003C\u002Fstrong>：从 API、PDF、SQL、文档等来源摄取数据，覆盖常见格式\u003C\u002Fli>\n\u003Cli>\u003Cstrong>索引构建\u003C\u002Fstrong>：把数据组织成向量索引、关键词索引等多种结构，供检索使用\u003C\u002Fli>\n\u003Cli>\u003Cstrong>查询引擎\u003C\u002Fstrong>：针对提问自动检索相关片段并组织回答\u003C\u002Fli>\n\u003Cli>\u003Cstrong>RAG 流水线\u003C\u002Fstrong>：从摄取、切分、嵌入到生成的完整链路开箱即用\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Agent 编排\u003C\u002Fstrong>：基于 Workflows 构建单 Agent 与多 Agent 应用\u003C\u002Fli>\n\u003Cli>\u003Cstrong>模块化集成\u003C\u002Fstrong>：LLM、嵌入模型、向量库均以独立集成包接入，替换成本低\u003C\u002Fli>\n\u003Cli>\u003Cstrong>结构化输出\u003C\u002Fstrong>：支持把回答约束为 JSON 等固定结构\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"适用人群\">适用人群\u003C\u002Fh2>\u003Cul>\n\u003Cli>构建 RAG 知识问答系统的开发者\u003C\u002Fli>\n\u003Cli>需要把企业文档接入 LLM 的工程师\u003C\u002Fli>\n\u003Cli>开发多 Agent 应用的 AI 团队\u003C\u002Fli>\n\u003Cli>快速验证 LLM 应用原型的产品开发者\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"快速上手\">快速上手\u003C\u002Fh2>\u003Col>\n\u003Cli>安装整合包（含核心与常用集成）：\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Cdiv class=\"code-block\">\u003Cpre>\u003Ccode class=\"hljs language-bash\">pip install llama-index\u003C\u002Fcode>\u003C\u002Fpre>\u003C\u002Fdiv>\u003Col start=\"2\">\n\u003Cli>或只装核心包，再按需添加集成：\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Cdiv class=\"code-block\">\u003Cpre>\u003Ccode class=\"hljs language-bash\">pip install llama-index-core\u003C\u002Fcode>\u003C\u002Fpre>\u003C\u002Fdiv>\u003Col start=\"3\">\n\u003Cli>参照官方文档完成 LLM 与向量库配置，构建索引并发起查询\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Cp>包名带 \u003Ccode>core\u003C\u002Fcode> 前缀的导入来自核心包，不带 \u003Ccode>core\u003C\u002Fcode> 的来自集成包，例如：\u003C\u002Fp>\n\u003Cdiv class=\"code-block\">\u003Cpre>\u003Ccode class=\"hljs language-python\">from llama_index.core.llms import LLM\nfrom llama_index.llms.openai import OpenAI\u003C\u002Fcode>\u003C\u002Fpre>\u003C\u002Fdiv>\u003Ch2 id=\"核心优势\">核心优势\u003C\u002Fh2>\u003Cul>\n\u003Cli>\u003Cstrong>生态完整\u003C\u002Fstrong>：从数据摄取到生成回答的整条链路都有现成组件\u003C\u002Fli>\n\u003Cli>\u003Cstrong>供应商中立\u003C\u002Fstrong>：300+ 集成包可自由替换模型与存储后端\u003C\u002Fli>\n\u003Cli>\u003Cstrong>上手快\u003C\u002Fstrong>：整合包一条命令安装，几行代码跑通第一个问答\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"系统要求\">系统要求\u003C\u002Fh2>\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>项目\u003C\u002Fth>\n\u003Cth>要求\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>操作系统\u003C\u002Ftd>\n\u003Ctd>Linux、macOS、Windows\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd>语言环境\u003C\u002Ftd>\n\u003Ctd>Python 3.9+\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n\u003Ch2 id=\"注意事项\">注意事项\u003C\u002Fh2>\u003Cul>\n\u003Cli>\u003Cstrong>项目重心\u003C\u002Fstrong>：团队当前主推文档解析平台 LlamaParse，框架本身仍可用但更新以文档站为准\u003C\u002Fli>\n\u003Cli>\u003Cstrong>包结构\u003C\u002Fstrong>：\u003Ccode>llama-index\u003C\u002Fcode> 整合包与 \u003Ccode>llama-index-core\u003C\u002Fcode> 核心包二选一，集成包按需另行安装\u003C\u002Fli>\n\u003Cli>\u003Cstrong>文档时效\u003C\u002Fstrong>：README 更新不及文档站频繁，以官方文档为准\u003C\u002Fli>\n\u003C\u002Ful>\n",1793197937623]