<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>RAG | Eyte Channel</title><description/><link>https://channel.nostses.icu</link><item><title>Retrieval-Augmented Generation (RAG) Project：这份开源资源帮你从零开始全面掌握RAG（检索增强生成）技术，实用且系统</title><link>https://channel.nostses.icu/posts/2209</link><guid isPermaLink="true">https://channel.nostses.icu/posts/2209</guid><pubDate>Thu, 22 Jan 2026 18:53:01 GMT</pubDate><content:encoded>&lt;div class=&quot;tgme_widget_message_forwarded_from accent_color&quot;&gt;Forwarded from &lt;a class=&quot;tgme_widget_message_forwarded_from_name&quot; href=&quot;https://t.me/HardcoreOpenAI/1051&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span&gt;硬核开源智库&lt;/span&gt;&lt;/a&gt;&lt;/div&gt;&lt;a href=&quot;https://github.com/bRAGAI/bRAG-langchain/&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Retrieval-Augmented Generation (RAG) Project&quot;&gt;Retrieval-Augmented Generation (&lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;) Project&lt;/a&gt;：这份开源资源帮你从零开始全面掌握RAG（检索增强生成）技术，实用且系统。&lt;br /&gt;&lt;br /&gt;主要内容涵盖：&lt;br /&gt;- 查询构建：将自然语言转成结构化查询（SQL、Cypher、向量检索）  &lt;br /&gt;- 查询翻译：分解、重构输入，提升检索效果  &lt;br /&gt;- 路由选择：动态选库或嵌入上下文，精准定位答案  &lt;br /&gt;- 检索优化：多种重排序算法+实时数据接入，确保结果相关性  &lt;br /&gt;- 索引管理：多重表征嵌入、分层摘要、结构化搜索提升效率  &lt;br /&gt;- 生成环节：自研Self-RAG和RRR，实现推理与检索的迭代闭环  &lt;br /&gt;&lt;br /&gt;每个笔记本都有详细的实操指导，适合入门到进阶，支持多查询、多模态等高级用法。&lt;br /&gt;&lt;br /&gt;如果你从事机器学习、LLM或AI代理，强烈推荐收藏并实践。本资源极大降低了构建复杂RAG应用的门槛，助你快速搭建高效智能系统。&lt;br /&gt;&lt;br /&gt;RAG的核心难题不只是架构，更是优质数据的积累与语料空白的补充。未来，递归推理与动态语料更新将成为关键突破点。&lt;br /&gt;&lt;a href=&quot;/search/result?q=%23%E8%B5%84%E6%BA%90%E5%8F%82%E8%80%83&quot; title=&quot;#资源参考&quot;&gt;#资源参考&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23RAG&quot; title=&quot;#RAG&quot;&gt;#RAG&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23%E5%BC%80%E6%BA%90RAG&quot; title=&quot;#开源RAG&quot;&gt;#开源RAG&lt;/a&gt;&lt;a class=&quot;tgme_widget_message_link_preview&quot; href=&quot;https://github.com/bRAGAI/bRAG-langchain&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Everything you need to know to build your own RAG application - bragai/bRAG-langchain&quot;&gt;
  
  &lt;div class=&quot;link_preview_site_name accent_color&quot;&gt;GitHub&lt;/div&gt;
  &lt;img class=&quot;link_preview_image&quot; alt=&quot;GitHub - bragai/bRAG-langchain: Everything you need to know to build your own RAG application&quot; src=&quot;/static/https://cdn4.telesco.pe/file/qvwLbaDWmoqSZG8SlB0BPJOuPUKUVd5gQIxhIYTig-BwVApk88oPMslWFNJHMrccM-M2cccxiRzqWv1rFHylbLU1T1cZXy5Ls17hpC206SilPgcN_OocygdPKWWpWh_cYbHc4X82m9YtR4r1muIdT2UGjXHkRnjpidIxjVIKTGEFwxZ10f_cr7cvGm3aYX6GShuF1AevMJ1DaeFuOCGi6bHg0PFsLa13OESq1Wd8kdX4MLJuFzvHINGOox3_LoPxY9qUBVmVXbtN5UWezM2WRrt_1CTJeMMaAaGdf4oYIJAeSrOAKv9GKyn2dRIHKRFGIX0qUfU7qhE5omCdZ1UgKg.jpg&quot; width=&quot;1200&quot; height=&quot;630&quot; loading=&quot;eager&quot; /&gt;
  &lt;div class=&quot;link_preview_title&quot;&gt;GitHub - bragai/bRAG-langchain: Everything you need to know to build your own &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt; application&lt;/div&gt;
  &lt;div class=&quot;link_preview_description&quot;&gt;Everything you need to know to build your own &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt; application - bragai/bRAG-langchain&lt;/div&gt;
&lt;/a&gt;</content:encoded></item><item><title>LlamaFarm 是一个开源框架，专注于构建基于检索增强（RAG）和智能代理的AI应用</title><link>https://channel.nostses.icu/posts/1907</link><guid isPermaLink="true">https://channel.nostses.icu/posts/1907</guid><pubDate>Sun, 28 Dec 2025 01:32:40 GMT</pubDate><content:encoded>&lt;div class=&quot;tgme_widget_message_forwarded_from accent_color&quot;&gt;Forwarded from &lt;a class=&quot;tgme_widget_message_forwarded_from_name&quot; href=&quot;https://t.me/HardcoreOpenAI/1020&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span&gt;硬核开源智库&lt;/span&gt;&lt;/a&gt;&lt;/div&gt;&lt;a href=&quot;https://github.com/llama-farm/llamafarm&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;LlamaFarm&quot;&gt;LlamaFarm&lt;/a&gt; 是一个开源框架，专注于构建基于检索增强（&lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;）和智能代理的AI应用。它内置了默认方案（本地模型 Ollama、向量存储 Chroma），但架构完全可扩展，支持随时替换运行时、数据库和解析器，无需重写代码。| &lt;a href=&quot;/search/result?q=%23%E6%A1%86%E6%9E%B6&quot; title=&quot;#框架&quot;&gt;#框架&lt;/a&gt;&lt;br /&gt;&lt;br /&gt;主要特点：&lt;br /&gt;- 本地优先体验，一条命令行工具（lf）管理项目、数据集和对话&lt;br /&gt;- 生产级架构，支持配置驱动、模式校验的项目管理&lt;br /&gt;- 灵活定制的RAG流水线，YAML配置即可轻松调整&lt;br /&gt;- 集成多种AI运行时，支持本地模型和云端API无缝切换&lt;br /&gt;- 丰富的CLI命令，实现项目初始化、数据上传、处理及聊天&lt;br /&gt;- 提供OpenAI格式兼容的REST API，方便集成到各种应用&lt;br /&gt;&lt;br /&gt;支持 macOS、Linux 和 Windows，安装便捷，助力开发者快速搭建强大AI系统。&lt;br /&gt;&lt;a href=&quot;/search/result?q=%23%E8%B5%84%E6%BA%90%E5%8F%82%E8%80%83&quot; title=&quot;#资源参考&quot;&gt;#资源参考&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23%E5%B7%A5%E5%85%B7&quot; title=&quot;#工具&quot;&gt;#工具&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23RAG&quot; title=&quot;#RAG&quot;&gt;#RAG&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23LLM&quot; title=&quot;#LLM&quot;&gt;#LLM&lt;/a&gt;&lt;a class=&quot;tgme_widget_message_link_preview&quot; href=&quot;https://github.com/llama-farm/llamafarm?utm_source=tldrai&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Deploy any AI model, agent, database, RAG, and pipeline locally or remotely in minutes - llama-farm/llamafarm&quot;&gt;
  
  &lt;div class=&quot;link_preview_site_name accent_color&quot;&gt;GitHub&lt;/div&gt;
  &lt;img class=&quot;link_preview_image&quot; alt=&quot;GitHub - llama-farm/llamafarm: Deploy any AI model, agent, database, RAG, and pipeline locally or remotely in minutes&quot; src=&quot;/static/https://cdn4.telesco.pe/file/H0_MVmeUn2WaYRvj29TTT3iPCtDQl7ZP7-uTZksaPrNYaK7CLFYdlMM1LHaOQcDVSN8agSaq07Ac7JNUrisl0So_rtq2vRki9dnJLWAE3nqD0mJvBdWmE7zaF5YLoaIcAPITdQLr3u2FW6alkfDDyz4xS6j8R6HZYJqyvX1EPB1mRJom-875kkZD6x5axxgXuntgbur0fETUBhaI4370-F3T2_BouPor4OcUSLDkbAZzldofCOvD5TT2l8S7Uk1rRlpNgKvp4EKnDI-U8naKNLBk3zxXUvGCjppM_7r_tPzaIYB0NUe3CgLm5bFWaiWoq8yOlU86qOAwsarr4pIJ6g.jpg&quot; width=&quot;1200&quot; height=&quot;630&quot; loading=&quot;eager&quot; /&gt;
  &lt;div class=&quot;link_preview_title&quot;&gt;GitHub - llama-farm/llamafarm: Deploy any AI model, agent, database, &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;, and pipeline locally or remotely in minutes&lt;/div&gt;
  &lt;div class=&quot;link_preview_description&quot;&gt;Deploy any AI model, agent, database, &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;, and pipeline locally or remotely in minutes - llama-farm/llamafarm&lt;/div&gt;
&lt;/a&gt;</content:encoded></item><item><title>AI工程不断迭代升级，想掌握LLM、RAG和智能代理的实战技巧？  AI Engineering Hub 是一个集深度教程与实战案例于一体的开源项目，覆盖大语言模型、检索增强生成、AI代理等前沿内容</title><link>https://channel.nostses.icu/posts/1899</link><guid isPermaLink="true">https://channel.nostses.icu/posts/1899</guid><pubDate>Sun, 28 Dec 2025 01:32:40 GMT</pubDate><content:encoded>&lt;div class=&quot;tgme_widget_message_forwarded_from accent_color&quot;&gt;Forwarded from &lt;a class=&quot;tgme_widget_message_forwarded_from_name&quot; href=&quot;https://t.me/HardcoreOpenAI/1005&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span&gt;硬核开源智库&lt;/span&gt;&lt;/a&gt;&lt;/div&gt;AI工程不断迭代升级，想掌握LLM、RAG和智能代理的实战技巧？  &lt;br /&gt;&lt;br /&gt;&lt;a href=&quot;https://github.com/patchy631/ai-engineering-hub&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;AI Engineering Hub&quot;&gt;AI Engineering Hub&lt;/a&gt; 是一个集深度教程与实战案例于一体的开源项目，覆盖大语言模型、检索增强生成、AI代理等前沿内容。无论你是入门者、开发者还是研究者，都能在这里找到丰富资源，助力项目落地和技能提升。&lt;br /&gt;&lt;br /&gt;主要特色：  &lt;br /&gt;- 系统讲解大型语言模型（LLM）和检索增强生成（&lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;）技术  &lt;br /&gt;- 丰富的AI智能代理实战案例，展示真实业务应用  &lt;br /&gt;- 详细示例代码，方便快速上手和二次开发  &lt;br /&gt;- 免费数据科学电子书赠送，涵盖150+核心课程，订阅即得&lt;br /&gt;&lt;a href=&quot;/search/result?q=%23%E8%B5%84%E6%BA%90%E5%8F%82%E8%80%83&quot; title=&quot;#资源参考&quot;&gt;#资源参考&lt;/a&gt; #&lt;a href=&quot;https://github.com/patchy631/ai-engineering-hub&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;AI Engineering Hub&quot;&gt;AI Engineering Hub&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23LLM&quot; title=&quot;#LLM&quot;&gt;#LLM&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23RAG&quot; title=&quot;#RAG&quot;&gt;#RAG&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23AI%E5%89%8D%E6%B2%BF%E5%86%85%E5%AE%B9&quot; title=&quot;#AI前沿内容&quot;&gt;#AI前沿内容&lt;/a&gt;&lt;a class=&quot;tgme_widget_message_link_preview&quot; href=&quot;https://github.com/patchy631/ai-engineering-hub&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;In-depth tutorials on LLMs, RAGs and real-world AI agent applications. - patchy631/ai-engineering-hub&quot;&gt;
  
  &lt;div class=&quot;link_preview_site_name accent_color&quot;&gt;GitHub&lt;/div&gt;
  &lt;img class=&quot;link_preview_image&quot; alt=&quot;GitHub - patchy631/ai-engineering-hub: In-depth tutorials on LLMs, RAGs and real-world AI agent applications.&quot; src=&quot;/static/https://cdn4.telesco.pe/file/Gwid5AJblIsyQu1ARi4KkoYwULAfJsJStbza4rDDs_pXenQaYmKQ-qQetxp6ZNhhL7j09QYLNiElE7uH090mfA3UjSNvSxUWrVeM8oAQz4a66qn_HC3XqG6ORrJKhfiqvj4U6MwMlgWHFRWO2em_hEAA-GexgRtOq4rYxOBYzWtazX0AURXur0Ryr_JAAmxIo0b4l0Jg_F8bAVUvINq1zB8Wxh586CAxUvCNMGrMalfNnXOIKjKv64oM46nK_DVts5t6Ds2t2Q767tQ3OqRjVWq1n1uYhihwN_53yRJDZB9sn9WHwIB5aYyg6pg5j-dDResKP5hXjnPK3VU5KDrNHg.jpg&quot; width=&quot;1200&quot; height=&quot;630&quot; loading=&quot;eager&quot; /&gt;
  &lt;div class=&quot;link_preview_title&quot;&gt;GitHub - patchy631/ai-engineering-hub: In-depth tutorials on LLMs, &lt;mark class=&quot;highlight&quot;&gt;RAGs&lt;/mark&gt; and real-world AI agent applications.&lt;/div&gt;
  &lt;div class=&quot;link_preview_description&quot;&gt;In-depth tutorials on LLMs, &lt;mark class=&quot;highlight&quot;&gt;RAGs&lt;/mark&gt; and real-world AI agent applications. - patchy631/ai-engineering-hub&lt;/div&gt;
&lt;/a&gt;</content:encoded></item><item><title>在线智能问答系统，文档上传、向量检索、模型推理一体化，轻松实现本地RAG入门体验</title><link>https://channel.nostses.icu/posts/1890</link><guid isPermaLink="true">https://channel.nostses.icu/posts/1890</guid><pubDate>Sun, 28 Dec 2025 01:32:39 GMT</pubDate><content:encoded>&lt;div class=&quot;tgme_widget_message_forwarded_from accent_color&quot;&gt;Forwarded from &lt;a class=&quot;tgme_widget_message_forwarded_from_name&quot; href=&quot;https://t.me/HardcoreOpenAI/988&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span&gt;硬核开源智库&lt;/span&gt;&lt;/a&gt;&lt;/div&gt;在线智能问答系统，文档上传、向量检索、模型推理一体化，轻松实现本地RAG入门体验。&lt;br /&gt;&lt;br /&gt;&lt;a href=&quot;https://github.com/weiwill88/Local_Pdf_Chat_RAG&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Local PDF Chat RAG&quot;&gt;Local PDF Chat &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;&lt;/a&gt; 是一个开源项目，专为想理解并动手实践检索增强生成（&lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;）技术的初学者设计。它结合了PDF处理、FAISS向量检索、多模型集成等核心模块，帮助你从源码层面掌握RAG的底层流程。&lt;br /&gt;&lt;br /&gt;主要功能：&lt;br /&gt;&lt;br /&gt;- 多PDF文档上传与自动文本切割向量化&lt;br /&gt;- 本地FAISS向量数据库构建与高效语义检索&lt;br /&gt;- 混合BM25关键词检索提升召回率&lt;br /&gt;- 交叉编码器及大模型（支持本地 Ollama 和云端 SiliconFlow）结果重排序&lt;br /&gt;- 支持联网搜索增强回答的时效性（需配置SerpAPI密钥）&lt;br /&gt;- 递归式深度检索，自动生成新查询，提升答复深度&lt;br /&gt;- Gradio交互式Web UI，操作简单直观&lt;br /&gt;- 本地化优先，保护数据隐私&lt;br /&gt;&lt;br /&gt;适合科研、开发者和RAG技术爱好者快速上手，理解RAG的全流程细节。&lt;br /&gt;&lt;a href=&quot;/search/result?q=%23%E8%B5%84%E6%BA%90%E5%8F%82%E8%80%83&quot; title=&quot;#资源参考&quot;&gt;#资源参考&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23%E5%B7%A5%E5%85%B7&quot; title=&quot;#工具&quot;&gt;#工具&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23RAG&quot; title=&quot;#RAG&quot;&gt;#RAG&lt;/a&gt;&lt;a class=&quot;tgme_widget_message_link_preview&quot; href=&quot;https://github.com/weiwill88/Local_Pdf_Chat_RAG&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;🧠 纯原生 Python 实现的 RAG 框架 | FAISS + BM25 混合检索 | 支持 Ollama / SiliconFlow | 适合新手入门学习 - weiwill88/Local_Pdf_Chat_RAG&quot;&gt;
  
  &lt;div class=&quot;link_preview_site_name accent_color&quot;&gt;GitHub&lt;/div&gt;
  &lt;img class=&quot;link_preview_image&quot; alt=&quot;GitHub - weiwill88/Local_Pdf_Chat_RAG: 🧠 纯原生 Python 实现的 RAG 框架 | FAISS + BM25 混合检索 | 支持 Ollama / SiliconFlow | 适合新手入门学习&quot; src=&quot;/static/https://cdn4.telesco.pe/file/Rq0gz2-ZgahrZTaPix8tb26qjFUhreaes-oll-zVnUHmIt1SSIGfGi1BAItinCeT8sHh63DZbsdxpKgCEJ3Jz7_IHYPKYUhze-rqVBdF4hxKRN4fyadJcFsmmMor-hKBC3ingp3qkk75SNqFHKdCvGHqL_mnKFPA8k7GMQiGAakSLdfD8EemARZ-h9txnq_2lAyAeSJ1CQ9iUTvDy6c_vwBtgGaseCNnQAPxVS6FcWqdtZgZhDMIv7VvfcqjV1Id0Bkx37lLzLmLiJ8JW_U0oYXLkRhnqVqLaeY_Miuj8IUb9HewpmHUJLET2bG4JL0tpCCikdotsuaoNY8IbRn0Gw.jpg&quot; width=&quot;1200&quot; height=&quot;630&quot; loading=&quot;eager&quot; /&gt;
  &lt;div class=&quot;link_preview_title&quot;&gt;GitHub - weiwill88/Local_Pdf_Chat_RAG: &lt;i class=&quot;emoji&quot; style=&quot;background-image:url(&apos;//telegram.org/img/emoji/40/F09FA7A0.png&apos;)&quot;&gt;&lt;b&gt;🧠&lt;/b&gt;&lt;/i&gt; 纯原生 Python 实现的 &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt; 框架 | FAISS + BM25 混合检索 | 支持 Ollama / SiliconFlow | 适合新手入门学习&lt;/div&gt;
  &lt;div class=&quot;link_preview_description&quot;&gt;&lt;i class=&quot;emoji&quot; style=&quot;background-image:url(&apos;//telegram.org/img/emoji/40/F09FA7A0.png&apos;)&quot;&gt;&lt;b&gt;🧠&lt;/b&gt;&lt;/i&gt; 纯原生 Python 实现的 &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt; 框架 | FAISS + BM25 混合检索 | 支持 Ollama / SiliconFlow | 适合新手入门学习 - weiwill88/Local_Pdf_Chat_RAG&lt;/div&gt;
&lt;/a&gt;</content:encoded></item><item><title>Ragflow-Plus：基于 Ragflow 的二次开发，专注解决实际应用痛点，提升知识库管理与文档交互效率</title><link>https://channel.nostses.icu/posts/1225</link><guid isPermaLink="true">https://channel.nostses.icu/posts/1225</guid><pubDate>Wed, 30 Jul 2025 17:46:08 GMT</pubDate><content:encoded>&lt;div class=&quot;tgme_widget_message_forwarded_from accent_color&quot;&gt;Forwarded from &lt;a class=&quot;tgme_widget_message_forwarded_from_name&quot; href=&quot;https://t.me/HardcoreOpenAI/702&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span&gt;硬核开源智库&lt;/span&gt;&lt;/a&gt;&lt;/div&gt;&lt;a href=&quot;https://github.com/zstar1003/ragflow-plus&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Ragflow-Plus&quot;&gt;Ragflow-Plus&lt;/a&gt;：基于 Ragflow 的二次开发，专注解决实际应用痛点，提升知识库管理与文档交互效率。&lt;br /&gt;&lt;br /&gt;• 全新后台管理系统：支持用户、团队、配置、文件与知识库统一管理，提升运维便捷度  &lt;br /&gt;• 权限回收机制：前端权限收缩，简化操作界面，保障安全与易用并重  &lt;br /&gt;• 解析能力升级：引入 MinerU 替代 DeepDoc，增强文本与图片解析效果，实现图文结合输出  &lt;br /&gt;• 文档撰写模式革新：全新交互体验，优化内容创作流程，适合多场景知识沉淀  &lt;br /&gt;• 开箱即用：提供 Docker 快速部署方案，配套视频教程与详细文档支持  &lt;br /&gt;• 开源透明：遵循 AGPLv3 许可证，支持商业使用，保障软件自由与合规  &lt;br /&gt;• 社群活跃：官方社群讨论技术与使用，支持持续贡献与共建  &lt;br /&gt;&lt;br /&gt;Ragflow-Plus 深耕知识管理本质，结合技术迭代与用户需求，助力打造高效、灵活的智能文档与问答平台。&lt;br /&gt;&lt;a href=&quot;/search/result?q=%23%E8%B5%84%E6%BA%90%E5%8F%82%E8%80%83&quot; title=&quot;#资源参考&quot;&gt;#资源参考&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23RAG&quot; title=&quot;#RAG&quot;&gt;#RAG&lt;/a&gt;&lt;a class=&quot;tgme_widget_message_link_preview&quot; href=&quot;https://github.com/zstar1003/ragflow-plus&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Ragflow-Plus 是 Ragflow 的二次开发版本，使其更为简洁实用. Contribute to zstar1003/ragflow-plus development by creating an account on GitHub.&quot;&gt;
  
  &lt;div class=&quot;link_preview_site_name accent_color&quot;&gt;GitHub&lt;/div&gt;
  
  &lt;div class=&quot;link_preview_title&quot;&gt;GitHub - zstar1003/ragflow-plus: Ragflow-Plus 是 Ragflow 的二次开发版本，使其更为简洁实用&lt;/div&gt;
  &lt;div class=&quot;link_preview_description&quot;&gt;Ragflow-Plus 是 Ragflow 的二次开发版本，使其更为简洁实用. Contribute to zstar1003/ragflow-plus development by creating an account on GitHub.&lt;/div&gt;
&lt;/a&gt;</content:encoded></item><item><title>Colette：面向技术文档的本地多模态检索增强生成（RAG）开源平台  • 核心采用视觉RAG（V-RAG）技术，将文档转为图像处理，完整保留图表、布局等视觉元素，提升对复杂技术文档的理解能力• 支持文本RAG，结合非结构化文本抽取、嵌入和主流大语言模型，实现多模态融合检索与交互  • 多模型支持，兼容多种嵌入器与视觉语言模型，灵活适配不同场景  • 集成图像生成（diffusers），增强交互体验与内容创作能力• 自托管部署，基于Docker，满足数据隐私需求，适合存储和处理敏感技术资料  • 适用环境配置明确（GPU≥24GB，内存≥16GB，磁盘≥50GB），确保性能稳定  • 详细命令行与Python API示例，方便快速集成与二次开发  • 困难排查指南助力优化检索准确性，支持社区反馈与持续迭代  从本质看，Colette围绕“视觉优先”的多模态理解方法，突破传统文本检索局限，提升技术文档智能交互的深度和精度，适合企业与研发机构构建安全、可控的知识管理系统</title><link>https://channel.nostses.icu/posts/1200</link><guid isPermaLink="true">https://channel.nostses.icu/posts/1200</guid><pubDate>Wed, 30 Jul 2025 17:35:38 GMT</pubDate><content:encoded>&lt;div class=&quot;tgme_widget_message_forwarded_from accent_color&quot;&gt;Forwarded from &lt;a class=&quot;tgme_widget_message_forwarded_from_name&quot; href=&quot;https://t.me/HardcoreOpenAI/631&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span&gt;硬核开源智库&lt;/span&gt;&lt;/a&gt;&lt;/div&gt;&lt;a href=&quot;http://github.com/jolibrain/colette&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Colette&quot;&gt;Colette&lt;/a&gt;：面向技术文档的本地多模态检索增强生成（&lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;）开源平台  &lt;br /&gt;&lt;br /&gt;• 核心采用视觉RAG（V-&lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;）技术，将文档转为图像处理，完整保留图表、布局等视觉元素，提升对复杂技术文档的理解能力&lt;br /&gt;• 支持文本RAG，结合非结构化文本抽取、嵌入和主流大语言模型，实现多模态融合检索与交互  &lt;br /&gt;• 多模型支持，兼容多种嵌入器与视觉语言模型，灵活适配不同场景  &lt;br /&gt;• 集成图像生成（diffusers），增强交互体验与内容创作能力&lt;br /&gt;• 自托管部署，基于Docker，满足数据隐私需求，适合存储和处理敏感技术资料  &lt;br /&gt;• 适用环境配置明确（GPU≥24GB，内存≥16GB，磁盘≥50GB），确保性能稳定  &lt;br /&gt;• 详细命令行与Python API示例，方便快速集成与二次开发  &lt;br /&gt;• 困难排查指南助力优化检索准确性，支持社区反馈与持续迭代  &lt;br /&gt;&lt;br /&gt;从本质看，Colette围绕“视觉优先”的多模态理解方法，突破传统文本检索局限，提升技术文档智能交互的深度和精度，适合企业与研发机构构建安全、可控的知识管理系统。&lt;br /&gt;&lt;a href=&quot;/search/result?q=%23%E8%B5%84%E6%BA%90%E5%8F%82%E8%80%83&quot; title=&quot;#资源参考&quot;&gt;#资源参考&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23RAG&quot; title=&quot;#RAG&quot;&gt;#RAG&lt;/a&gt;&lt;a class=&quot;tgme_widget_message_link_preview&quot; href=&quot;https://github.com/jolibrain/colette&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;Multimodal RAG to search and interact locally with technical documents of any kind - jolibrain/colette&quot;&gt;
  
  &lt;div class=&quot;link_preview_site_name accent_color&quot;&gt;GitHub&lt;/div&gt;
  &lt;img class=&quot;link_preview_image&quot; alt=&quot;GitHub - jolibrain/colette: Multimodal RAG to search and interact locally with technical documents of any kind&quot; src=&quot;/static/https://cdn4.telesco.pe/file/R-xuwcdl_6zLGgHar086ca7p7EEeYks43HvdvYR-ybmOebG8uWrl0RfdXryW1im64-mCG6oKptvaersXbfWSXIZT6M0JsJbl_q-1ShndXXtuMUi5-9CDbQSfi192ystDGwrbYXT88du42FNMK1yV0sjpL5xv83jLzoRZBYoIXNEqrRKlSZbSTfvy2RijL2ahXTGocmpvfRxGOSdfp2gQR9_rBNSIS0-FEvoBznp_97_kEvJAx0Wt5DJ5lH1N-ktDH-v2YEaGDW3wlTne4Q02pPqn4-SWSB9okfWewtUusEROL7HlExeBn7XCDGyF1JbiiB12wf5Znyx1tuQoauFj5Q.jpg&quot; width=&quot;1200&quot; height=&quot;630&quot; loading=&quot;eager&quot; /&gt;
  &lt;div class=&quot;link_preview_title&quot;&gt;GitHub - jolibrain/colette: Multimodal &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt; to search and interact locally with technical documents of any kind&lt;/div&gt;
  &lt;div class=&quot;link_preview_description&quot;&gt;Multimodal &lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt; to search and interact locally with technical documents of any kind - jolibrain/colette&lt;/div&gt;
&lt;/a&gt;</content:encoded></item><item><title>Simba 是一个开源知识管理系统，旨在与任何检索增强生成 (RAG) 系统无缝集成</title><link>https://channel.nostses.icu/posts/1064</link><guid isPermaLink="true">https://channel.nostses.icu/posts/1064</guid><pubDate>Thu, 10 Jul 2025 05:27:35 GMT</pubDate><content:encoded>&lt;div class=&quot;tgme_widget_message_forwarded_from accent_color&quot;&gt;Forwarded from &lt;a class=&quot;tgme_widget_message_forwarded_from_name&quot; href=&quot;https://t.me/HardcoreOpenAI/491&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot;&gt;&lt;span&gt;硬核开源智库&lt;/span&gt;&lt;/a&gt;&lt;/div&gt;Simba 是一个开源知识管理系统，旨在与任何检索增强生成 (&lt;mark class=&quot;highlight&quot;&gt;RAG&lt;/mark&gt;) 系统无缝集成。&lt;br /&gt;&lt;br /&gt;借助现代化的 UI 和模块化架构，开发人员可以专注于构建人工智能解决方案，而不必担心知识管理的复杂性。&lt;br /&gt;&lt;br /&gt;&lt;i class=&quot;emoji&quot;&gt;&lt;b&gt;🧬&lt;/b&gt;&lt;/i&gt; &lt;a href=&quot;https://github.com/GitHamza0206/simba&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;https://github.com/GitHamza0206/simba&quot;&gt;https://github.com/GitHamza0206/simba&lt;/a&gt;&lt;br /&gt;&lt;a href=&quot;/search/result?q=%23%E8%B5%84%E6%BA%90%E5%8F%82%E8%80%83&quot; title=&quot;#资源参考&quot;&gt;#资源参考&lt;/a&gt; &lt;a href=&quot;/search/result?q=%23RAG&quot; title=&quot;#RAG&quot;&gt;#RAG&lt;/a&gt;&lt;a class=&quot;tgme_widget_message_link_preview&quot; href=&quot;https://github.com/GitHamza0206/simba&quot; target=&quot;_blank&quot; rel=&quot;noopener&quot; title=&quot;OpenSource Production ready Customer service with built in Evals and monitoring  - GitHamza0206/simba&quot;&gt;
  
  &lt;div class=&quot;link_preview_site_name accent_color&quot;&gt;GitHub&lt;/div&gt;
  &lt;img class=&quot;link_preview_image&quot; alt=&quot;GitHub - GitHamza0206/simba: OpenSource Production ready Customer service with built in Evals and monitoring&quot; src=&quot;/static/https://cdn4.telesco.pe/file/EZmeQGSTvy93-UXT_QRGc83g872YHIcKaA5UogeU9-cSzXnW05STyrMvvUsrCpAUIYMQ8UAHMFqTvuFBQowoSatx0pPF8trZz0pmfGcu8_94WfZmOI5Hj0IpnX-l_4wQMWs7ZG4GXQ2N-f3TRzrexWxyYUKiLMxFW9pGdOjtqZD0OiO3gUomW4M3E8iNt39IXOYsGWO5q5yH9CfN3LVFkdrPZ_Fwl62yFHLyp7h4-thz9-opLfFQNDUVS1YJp0clUVF-Q-fLICieeVmOZZmvcuoN8cbG9huns5czdHSOCbVcoQFb9QgvA_jUcIlDioMX6fKsLizRNpMHZ49-VC4VFQ.jpg&quot; width=&quot;1200&quot; height=&quot;630&quot; loading=&quot;eager&quot; /&gt;
  &lt;div class=&quot;link_preview_title&quot;&gt;GitHub - GitHamza0206/simba: OpenSource Production ready Customer service with built in Evals and monitoring&lt;/div&gt;
  &lt;div class=&quot;link_preview_description&quot;&gt;OpenSource Production ready Customer service with built in Evals and monitoring  - GitHamza0206/simba&lt;/div&gt;
&lt;/a&gt;</content:encoded></item></channel></rss>