[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-ragflow-open-source-rag-agent-engine-en":3,"tags-ragflow-open-source-rag-agent-engine-en":37,"related-lang-ragflow-open-source-rag-agent-engine-en":46,"related-posts-ragflow-open-source-rag-agent-engine-en":50,"series-tools-2834efcd-6c3d-424b-8436-68dbfade265b":87},{"id":4,"title":5,"content":6,"summary":7,"source":8,"source_url":9,"author":10,"image_url":11,"keywords":12,"language":18,"translated_content":10,"views":19,"is_premium":20,"created_at":21,"updated_at":21,"cover_image":11,"published_at":22,"rewrite_status":23,"rewrite_error":10,"rewritten_from_id":24,"slug":25,"category":26,"related_article_id":27,"status":28,"google_indexed_at":29,"x_posted_at":10,"tweet_text":10,"title_rewritten_at":10,"title_original":10,"key_takeaways":30,"topic_cluster_id":34,"embedding":35,"is_canonical_seed":36},"2834efcd-6c3d-424b-8436-68dbfade265b","RAGFlow adds agents to open-source RAG","\u003Cp data-speakable=\"summary\">RAGFlow is an open-source RAG engine that adds \u003Ca href=\"\u002Ftag\u002Fagent\">agent\u003C\u002Fa> features and self-hosting.\u003C\u002Fp>\u003Cp>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Finfiniflow\u002Fragflow\" target=\"_blank\" rel=\"noopener\">RAGFlow\u003C\u002Fa> is one of those projects that gets more interesting the more data you throw at it. The repository now shows \u003Cstrong>80.3k stars\u003C\u002Fstrong>, \u003Cstrong>9.1k forks\u003C\u002Fstrong>, and \u003Cstrong>6,155 commits\u003C\u002Fstrong>, which tells you this is not a side project tucked away in a corner.\u003C\u002Fp>\u003Cp>The pitch is simple: take retrieval-augmented generation, add agent capabilities, and make the whole thing usable for real document-heavy workflows. The repo also points people to \u003Ca href=\"https:\u002F\u002Fcloud.ragflow.io\" target=\"_blank\" rel=\"noopener\">RAGFlow Cloud\u003C\u002Fa> for a hosted option, while the open-source path stays centered on \u003Ca href=\"\u002Ftag\u002Fdocker\">Docker\u003C\u002Fa>, local setup, and a stack that can be tuned for different deployment sizes.\u003C\u002Fp>\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Fact\u003C\u002Fth>\u003Cth>Value\u003C\u002Fth>\u003Cth>Why it matters\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>GitHub stars\u003C\u002Ftd>\u003Ctd>80.3k\u003C\u002Ftd>\u003Ctd>Shows strong developer adoption\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>GitHub forks\u003C\u002Ftd>\u003Ctd>9.1k\u003C\u002Ftd>\u003Ctd>Signals active reuse and experimentation\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Commits\u003C\u002Ftd>\u003Ctd>6,155\u003C\u002Ftd>\u003Ctd>Points to steady product churn\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Minimum CPU\u003C\u002Ftd>\u003Ctd>4 cores\u003C\u002Ftd>\u003Ctd>Basic self-hosting floor\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Minimum RAM\u003C\u002Ftd>\u003Ctd>16 GB\u003C\u002Ftd>\u003Ctd>Needed for practical local deployment\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Minimum disk\u003C\u002Ftd>\u003Ctd>50 GB\u003C\u002Ftd>\u003Ctd>Storage budget for the full stack\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>What RAGFlow is trying to solve\u003C\u002Fh2>\u003Cp>Most RAG systems break down when the source material stops looking like clean Markdown. RAGFlow is built for the messier reality: PDFs with tables, scanned files, slide decks, spreadsheets, web pages, and mixed document stores that need grounding before a model can answer anything useful.\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778526659080-t2is.png\" alt=\"RAGFlow adds agents to open-source RAG\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>The project describes itself as a “converged context engine” with pre-built agent templates. In plain English, that means it tries to do more than simple chunk-and-embed retrieval. It wants to understand documents deeply, extract structure, and keep citations tied to the answer so humans can check the source.\u003C\u002Fp>\u003Cp>That matters because the failure mode in \u003Ca href=\"\u002Ftag\u002Fenterprise-ai\">enterprise AI\u003C\u002Fa> is usually not model quality alone. It is context quality. If the input layer is sloppy, the output layer looks confident while being wrong.\u003C\u002Fp>\u003Cul>\u003Cli>Deep document understanding for unstructured and mixed-format data\u003C\u002Fli>\u003Cli>Template-based chunking with visible, inspectable citation paths\u003C\u002Fli>\u003Cli>Support for Word, slides, Excel, text, images, scanned copies, structured data, and web pages\u003C\u002Fli>\u003Cli>Multiple recall paths with fused re-ranking for better answer selection\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>What changed in recent updates\u003C\u002Fh2>\u003Cp>The update log is unusually dense, and that is useful because it shows where the project is heading. Recent entries mention support for \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fdeepseek-ai\u002FDeepSeek-V4\" target=\"_blank\" rel=\"noopener\">DeepSeek v4\u003C\u002Fa>, \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fopenai\u002Fgpt-5\" target=\"_blank\" rel=\"noopener\">GPT-5\u003C\u002Fa>, and \u003Ca href=\"https:\u002F\u002Fdeepmind.google\u002Ftechnologies\u002Fgemini\u002F\" target=\"_blank\" rel=\"noopener\">Gemini 3 Pro\u003C\u002Fa>, along with agent memory, MCP, and data sync from Confluence, S3, Notion, Discord, and \u003Ca href=\"\u002Ftag\u002Fgoogle\">Google\u003C\u002Fa> Drive.\u003C\u002Fp>\u003Cblockquote>\"The most important thing about AI is that we need to understand that it is not magic.\" — \u003Ca href=\"https:\u002F\u002Fwww.nytimes.com\u002F2016\u002F12\u002F03\u002Fopinion\u002Fsundar-pichai-artificial-intelligence.html\" target=\"_blank\" rel=\"noopener\">Sundar Pichai\u003C\u002Fa>\u003C\u002Fblockquote>\u003Cp>That quote fits RAGFlow well. The project is not selling magic. It is trying to make the plumbing better, which is usually where enterprise AI either wins or fails. The newer features point to a system that is moving from document retrieval toward agentic workflows that can act on that context.\u003C\u002Fp>\u003Cp>Here are the recent updates that matter most if you are evaluating it for a team:\u003C\u002Fp>\u003Cul>\u003Cli>\u003Cstrong>2026-04-24:\u003C\u002Fstrong> DeepSeek v4 support\u003C\u002Fli>\u003Cli>\u003Cstrong>2026-03-24:\u003C\u002Fstrong> RAGFlow Skill on OpenClaw for dataset access\u003C\u002Fli>\u003Cli>\u003Cstrong>2025-12-26:\u003C\u002Fstrong> agent memory support\u003C\u002Fli>\u003Cli>\u003Cstrong>2025-11-19:\u003C\u002Fstrong> Gemini 3 Pro support\u003C\u002Fli>\u003Cli>\u003Cstrong>2025-11-12:\u003C\u002Fstrong> sync from Confluence, S3, Notion, Discord, and Google Drive\u003C\u002Fli>\u003Cli>\u003Cstrong>2025-10-23:\u003C\u002Fstrong> MinerU and Docling parsing methods\u003C\u002Fli>\u003Cli>\u003Cstrong>2025-10-15:\u003C\u002Fstrong> orchestrable ingestion pipeline\u003C\u002Fli>\u003Cli>\u003Cstrong>2025-08-08:\u003C\u002Fstrong> GPT-5 series support\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>How the self-hosted setup compares\u003C\u002Fh2>\u003Cp>RAGFlow is opinionated about deployment. The documentation asks for \u003Cstrong>4 CPU cores\u003C\u002Fstrong>, \u003Cstrong>16 GB RAM\u003C\u002Fstrong>, and \u003Cstrong>50 GB disk\u003C\u002Fstrong> as a baseline, plus \u003Cstrong>Docker 24.0.0\u003C\u002Fstrong> and \u003Cstrong>Docker Compose v2.26.1\u003C\u002Fstrong>. If you want code execution in its sandboxed agent flow, you also need \u003Ca href=\"https:\u002F\u002Fgvisor.dev\" target=\"_blank\" rel=\"noopener\">gVisor\u003C\u002Fa>.\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778526660696-txj3.png\" alt=\"RAGFlow adds agents to open-source RAG\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>That is a real-world setup, not a toy demo. The repo also warns that the published Docker images are built for x86, so ARM64 users need to build their own image. For teams running Macs with \u003Ca href=\"\u002Ftag\u002Fapple\">Apple\u003C\u002Fa> Silicon or ARM servers, that detail is easy to miss and painful to discover late.\u003C\u002Fp>\u003Cp>The storage engine choice is another practical signal. RAGFlow uses \u003Ca href=\"https:\u002F\u002Fwww.elastic.co\u002Felasticsearch\" target=\"_blank\" rel=\"noopener\">Elasticsearch\u003C\u002Fa> by default for full text and vectors, but it also supports switching to \u003Ca href=\"https:\u002F\u002Finfiniflow.org\u002Finfinity\" target=\"_blank\" rel=\"noopener\">Infinity\u003C\u002Fa>. That kind of flexibility matters when you are tuning cost, performance, and operational complexity.\u003C\u002Fp>\u003Cp>If you are comparing RAGFlow with a lighter RAG stack, the trade-off is clear: you get more document intelligence and more control, but you also accept more setup work. The project is aimed at teams that care about citations, multi-source ingestion, and production deployment rather than a quick weekend prototype.\u003C\u002Fp>\u003Cp>For readers who want related context, OraCore has also covered \u003Ca href=\"\u002Fnews\u002Fclaude-code-github-repo-analysis\" target=\"_blank\" rel=\"noopener\">Claude Code\u003C\u002Fa> and \u003Ca href=\"\u002Fnews\u002Fopenai-mcp-registry-explained\" target=\"_blank\" rel=\"noopener\">MCP registry workflows\u003C\u002Fa>.\u003C\u002Fp>\u003Ch2>Why this matters for builders\u003C\u002Fh2>\u003Cp>RAGFlow is interesting because it sits between a document pipeline and an application framework. The project is trying to reduce the gap between raw enterprise content and answers that can be trusted, cited, and acted on by agents.\u003C\u002Fp>\u003Cp>That positioning is timely. Teams are past the stage where they want a chatbot that can answer one-off questions from a PDF. They want systems that can ingest multiple sources, preserve traceability, and support workflows that mix retrieval, tool use, and memory.\u003C\u002Fp>\u003Cp>If RAGFlow keeps shipping at the current pace, the next question is not whether it can ingest more formats. It is whether teams will standardize on it as the context layer for internal AI products, or keep using it as a specialized tool for document-heavy workloads.\u003C\u002Fp>\u003Cp>For now, the clearest takeaway is practical: if your AI app depends on messy source data, RAGFlow is worth a serious look, especially if you need self-hosting and citation-aware retrieval. If your stack is already clean and simple, the extra machinery may be more than you need.\u003C\u002Fp>\u003Cp>What matters next is whether the project can keep balancing breadth, setup complexity, and reliability as agent features keep expanding.\u003C\u002Fp>","RAGFlow pairs retrieval-augmented generation with agent features, Docker self-hosting, and support for newer models like GPT-5 and Gemini 3 Pro.","github.com","https:\u002F\u002Fgithub.com\u002Finfiniflow\u002Fragflow",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778526659080-t2is.png",[13,14,15,16,17],"RAGFlow","RAG","agents","self-hosting","LLM context layer","en",3,false,"2026-05-11T19:10:36.388185+00:00","2026-05-11T19:10:36.364+00:00","done","07936925-f5fc-4fff-b003-9f6e7e420550","ragflow-open-source-rag-agent-engine-en","tools","e60e5a80-ca58-46fb-be85-5d51d7f3d2df","published","2026-05-12T09:00:13.342+00:00",[31,32,33],"RAGFlow combines retrieval-augmented generation with agent features for document-heavy AI apps.","The project has strong GitHub traction, with 80.3k stars, 9.1k forks, and 6,155 commits.","Recent updates add support for newer models, memory, MCP, and multiple enterprise data sources.","a04bc7fc-3b51-4bf4-8623-ded57ceaaf88","[-0.022342673,-0.0047739255,0.021985516,-0.08816036,-0.018443573,0.003530435,-0.019831993,0.019859208,0.005964058,0.02243383,0.0021224397,0.0031362344,0.008161767,-0.013958086,0.14629088,0.026636252,0.021239296,0.010292864,0.023636643,-0.024069335,-0.020002402,0.0023778623,-0.02165003,-0.009316026,0.014008332,0.0009346193,0.0040264935,-0.020567855,0.026278695,0.017333925,-0.011953687,-0.002775793,0.003739969,0.0020421993,0.012873578,0.024160938,0.025211917,-0.009317528,0.0063620126,0.012873461,-0.0109410705,-0.0046287123,0.007242091,-0.019988868,-0.019242555,-0.024725137,-0.01311887,-0.01169338,-0.0017866555,0.021743532,-0.035039797,0.018037695,0.012593818,-0.15425044,0.0055384208,0.016626196,0.008567714,0.010401765,0.019318746,0.026186733,0.019327948,-0.0022991542,-0.011512452,-0.016319469,-0.017748209,-0.009250252,0.0128107425,-0.010301246,-0.028033914,0.003363932,0.023469007,0.023622056,0.0038199422,-0.01823181,0.0026019064,-0.035549562,0.013681105,0.015614009,0.025543856,0.02530373,0.02309188,-0.009516925,-0.0052292296,0.0051619667,0.0004128189,-0.013703297,-0.013458144,-0.0157904,0.014910626,-0.003745933,0.021041445,-0.0039268592,0.034745317,0.034728706,-0.0067431447,-0.0150349485,0.004785705,-0.0047551785,-0.021131031,0.014165363,-0.011777776,-0.018958641,0.009216297,0.035495896,0.01311857,0.005760489,0.016889248,0.01967754,0.007205805,0.033063386,0.009194596,-0.0058331653,-0.021704135,0.0020336353,0.0066033644,-0.10387056,0.021488732,-4.8445712e-05,0.0042784647,-0.0063648275,-0.019989887,0.02305525,1.4545696e-05,0.01063741,0.021774914,-0.013598295,0.0013359722,0.0052070417,-0.016779402,0.04879869,-0.03560165,-0.0064685387,0.0065407897,0.006766875,-0.023169026,0.014098839,0.010832005,-0.0017473766,0.005614451,-0.020134464,0.0073345313,0.023358695,-0.0016290977,-0.037034944,-0.02317555,0.006479715,-0.01309198,0.020564659,0.0036236634,-0.01872054,0.01893364,0.008315489,0.008121222,-0.014118325,0.01933395,-0.011204864,-0.006264594,0.00073248293,-0.007616496,0.015783895,0.0026368108,-0.00011678125,0.022927362,0.010287058,0.008750312,0.016311642,0.013107359,0.01172182,0.01113876,0.029061068,-0.0163231,-0.012964763,0.006467184,-0.006554591,0.026966786,0.018239746,-0.0033930684,-0.0066074044,0.019611042,-0.035419688,0.008711144,-0.0030284568,-0.016439639,0.017197901,0.0041703926,0.0038277553,0.005235697,-0.0058036745,-0.003126809,-0.0055754823,-0.033670794,0.026751975,0.015429851,0.004701862,0.012566847,-0.0055923285,0.012415538,0.0076525616,0.010743253,0.02735358,0.015356661,-0.012812457,0.011568517,-0.026104702,-0.015541873,0.0117073525,-0.022561973,-0.018880809,0.00646659,-0.017671775,-0.0084435595,0.003413628,0.006648892,0.007385647,0.008833729,-0.018793901,0.013617444,-0.030723661,-0.003788295,-0.022309035,0.008343331,-0.011418118,0.024135944,-0.0011208091,-0.0040092906,-0.0031283936,-0.010315115,0.0012225905,-0.0019735019,0.019974707,0.00302576,0.025053024,-0.026159653,0.012813676,0.005833261,-0.006953098,-0.010275265,0.011540197,0.0061826017,0.009678687,-0.02535031,0.027596777,-0.008735918,0.015720308,0.027222114,-0.02465616,0.014922419,-0.013569767,-0.014770687,-0.010093536,-0.00933126,0.031574905,-0.025352959,-0.01859863,-0.023328146,-0.0007343306,0.00595447,0.009556066,-0.0141490465,-0.016129864,-0.007811241,0.015617416,-0.020166151,0.008843073,-0.010428761,0.023878464,0.026497066,-0.0064722374,-0.010150645,-0.0036886523,0.01085215,0.041816566,0.029358417,0.019225834,0.0061355773,-0.012887349,-0.022022929,0.046268817,-0.012254016,-0.0051425006,-0.014007144,0.012404244,-0.0026092865,-0.00024118791,0.021104498,0.0075725284,-0.010126046,-0.0070075793,-0.0060158456,0.018651992,0.0285958,0.034018118,-0.0065528196,-3.3280794e-06,0.003433327,-0.022818256,-0.02477066,0.0051062936,0.039237726,0.008842648,-0.0018964516,-0.012095829,0.016416995,0.029481102,-0.020965045,-0.0017176351,0.012356187,0.029065004,-0.027221765,0.0017688387,0.010706511,-0.012432936,-0.0057113734,-0.03393823,0.0022881047,-0.010864246,-0.009814177,-0.0136207985,-0.020973938,-0.006490322,-0.007261974,-0.0121455025,-0.016878491,-0.007922785,0.012058706,-0.029979035,-0.0004085284,0.013448721,0.026108239,0.009482926,0.013276903,-0.018475069,0.013657461,-0.0035778158,-0.0034719256,0.016960766,0.011174277,-0.016392073,0.0008191531,0.03969172,-0.01735808,0.0021031988,-0.014369283,-0.009982919,-0.007931832,0.02724686,0.027667388,-0.019781137,0.0038937335,-0.052331943,0.021718945,0.0030057835,-0.010385049,-0.04608571,-0.042655345,0.018791469,-0.024428047,0.011526969,0.03700303,0.0005325096,-0.007864137,0.021825364,-0.007730918,-0.015100423,0.003527028,-0.0030240952,-0.0145518435,0.013372705,-0.00019973269,0.01554111,0.011909482,-0.018741721,-0.0026150688,-0.0101804,0.004623314,0.0022025695,-0.0025869617,-0.0048248013,-0.03383403,0.018950062,-0.027581176,0.025339006,-0.021481771,-0.023449,0.0121583035,0.035503086,0.012563326,0.014660585,0.009425994,-0.014787285,-0.006387552,0.006775476,-0.018480983,0.008729884,0.010145163,-0.016085513,-0.02938863,0.019890675,-0.0036954188,-0.027442519,-0.016999269,0.028313475,-0.034725286,0.020494502,-0.0013287974,0.0014836434,-0.009319293,0.021481186,-0.017722076,-0.013292127,0.011839138,0.013812892,0.031144878,-0.006119968,0.021829229,-0.0031284639,0.00015668293,-0.004965247,0.008897831,0.0018905044,0.0016888754,0.023627862,-9.35533e-05,0.021334693,-0.025390295,-0.0022847126,-0.02406987,-0.010355748,0.0101715345,0.0023071764,0.020507563,-0.018814802,-0.0077879666,-0.026850345,-0.029008593,-0.01759488,-0.018604515,-0.024570657,-0.026122086,-0.0124828825,-0.0014418769,0.005816892,-0.024802543,0.011339124,0.018494248,0.008407336,-0.001101926,-0.011578111,-0.04940858,0.009852444,0.021897918,0.008524012,0.002728197,-0.009939927,0.00073580974,0.019398093,0.0014955046,0.01677795,0.004217913,-0.015576308,-0.007843156,0.0034293807,-0.008815336,-0.0124726,0.026766019,-0.02451212,-0.024771761,-0.031819012,0.00071775465,0.029802904,0.00023617508,0.03012532,0.022149803,-0.0013257938,0.011595995,-0.0010447949,0.009124715,0.006271023,-0.0015133137,-0.020322872,0.010814316,0.008725601,0.016213002,0.0072763856,0.01642561,0.02524891,-0.01357742,-0.00040190716,0.011003022,0.00079033023,0.031564165,-0.0025234222,-0.015422514,-0.008070794,-0.017165394,0.008155155,-0.011025143,0.00448018,0.0041072452,-0.0040030205,-0.010984767,-0.0025290458,0.00966881,-0.007418355,-0.025632054,-0.0059087337,-0.0043527232,-0.002991033,0.03533085,0.010637402,-0.03228942,0.019606669,-0.007560531,-0.006810917,-0.0073227123,-0.015180591,-0.010981432,0.0066540004,0.010883502,0.014799315,-0.009324948,0.0102991415,-0.031703282,0.008612591,-0.024546284,-0.031170838,0.01431658,-0.0051360554,0.00027276968,-0.00020133029,0.012610591,0.031977348,-0.0014444997,-7.003152e-05,0.009415167,-0.01489044,0.024961539,0.020423027,-0.017356582,0.02440556,-0.0076332767,0.0010505903,-0.0125107225,0.00035938565,0.041998178,-0.08897541,0.017730927,0.005391949,0.009653044,0.0015100392,0.0008734958,-0.0004896779,-0.015447662,-0.012489587,0.028810833,-0.0075265067,-0.01813983,0.029548073,0.012880702,-0.003036563,-0.001054626,0.001768035,-0.028181635,0.013882118,-0.0057934206,0.035058748,0.00018949337,0.0006252209,-0.0026388923,0.01576173,-0.024699528,0.016779937,-0.0064338585,0.036538668,0.011266207,-0.004606398,-0.005956889,-0.0024518636,0.006461374,0.010973965,0.009136825,0.010567274,-0.0052463347,0.010306162,-0.0053736586,0.01708049,0.011696515,0.003294039,-0.047292236,-0.023806222,0.005265501,-0.02827744,-0.009805997,0.018026398,-0.011953129,-0.04672649,-0.007236195,0.0012149013,-0.013415967,-0.0160934,-0.006938807,-0.019450402,0.01356773,-0.010432249,0.0028699972,-0.004504856,-0.036819622,-0.01928544,0.0064545576,0.0023081568,0.025785362,0.0012095451,0.024128959,0.003149685,0.024125485,0.019538088,-0.031022232,0.00024675432,-0.012054045,-0.020532042,0.012620036,-0.002640413,-0.0018704581,-0.0115019735,-0.012192598,-0.03819551,-0.013924512,-0.06624413,0.0019534768,-0.0019287523,-0.012831622,0.00029496645,0.006214978,-0.0053484314,-0.033641893,-0.011250002,-0.0052599213,0.029920774,0.010114971,-0.026327336,-0.016847746,0.023585558,0.033789154,0.015065944,0.037932225,0.038709097,-0.026332812,-0.011419593,-0.008803809,-0.010982102,-0.024552014,-0.023076836,0.002670535,-0.020967843,0.0013699032,-0.007018027,-0.017994354,0.012550912,-0.118266866,0.011020352,-0.011921186,-0.015476153,0.016713046,-0.0024423648,0.0050487993,0.037401754,-0.007922543,-0.018962968,0.018738486,-0.02448881,-0.012072097,-0.0065974826,0.0023424244,0.1300856,-0.0073028593,-0.016302207,-0.0076289745,-0.021815823,-0.0002030747,-0.022848347,0.011881318,0.011917165,0.00096442626,-0.01661322,0.014097803,-0.0173441,0.010671702,0.032470714,0.015605183,0.028544985,-0.011061251,-0.009400998,0.019682057,0.0071127503,0.0070246845,-0.02288311,0.018120397,0.0057102405,0.010888445,0.0030984844,-0.010031892,-0.036374237,0.03174841,-0.0073203784,-0.03511652,-0.02977194,-0.007945097,-0.023272857,-0.019293366,-0.10935402,0.0029161114,-0.0069447295,-0.0011863476,0.027337803,-0.017083503,0.037205786,0.02195378,0.0074079777,0.0018132802,-0.014785208,0.00064277917,0.0028915573,-0.019698467,0.029618505,-0.011592991,0.009924364,0.020615324,0.018976757,-0.008521531,0.0013173649,0.0062153535,0.00046415886,-0.009855618,-0.006315724,0.004226397,0.004572758,0.040870927,-0.023311356,0.0024722717,-0.017237648,0.017734941,-0.0064076954,0.014312533,0.005309426,0.0040796404,-0.031356305,0.00974533,-0.025472945,-0.010416814,0.010154119,0.0027215655,0.02809242,0.020804254,0.021022663,0.019545114,-0.0066471375,0.013078944,-0.008916274,-0.016131004,-0.037536137,0.031941127,-0.024054825,-0.013755339,0.020841336,0.013294165,0.010958021,0.03370044,0.002533204]",true,[38,40,41,42,44],{"name":14,"slug":39},"rag",{"name":15,"slug":15},{"name":16,"slug":16},{"name":13,"slug":43},"ragflow",{"name":17,"slug":45},"llm-context-layer",{"id":27,"slug":47,"title":48,"language":49},"ragflow-open-source-rag-agent-engine-zh","RAGFlow 加入 Agent 與自架部署","zh",[51,57,63,69,75,81],{"id":52,"slug":53,"title":54,"cover_image":55,"image_url":55,"created_at":56,"category":26},"8b02abfa-eb16-4853-8b15-63d302c7b587","why-vidhub-huiyuan-hutong-bushi-quan-shebei-tongyong-en","Why VidHub 会员互通不是“买一次全设备通用”","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778789439875-uceq.png","2026-05-14T20:10:26.046635+00:00",{"id":58,"slug":59,"title":60,"cover_image":61,"image_url":61,"created_at":62,"category":26},"abe54a57-7461-4659-b2a0-99918dfd2a33","why-buns-zig-to-rust-experiment-is-right-en","Why Bun’s Zig-to-Rust experiment is the right move","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778767895201-5745.png","2026-05-14T14:10:29.298057+00:00",{"id":64,"slug":65,"title":66,"cover_image":67,"image_url":67,"created_at":68,"category":26},"f0015918-251b-43d7-95af-032d2139f3f6","why-openai-api-pricing-is-product-strategy-en","Why OpenAI API pricing is a product strategy, not a footnote","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778749841805-uyhg.png","2026-05-14T09:10:27.921211+00:00",{"id":70,"slug":71,"title":72,"cover_image":73,"image_url":73,"created_at":74,"category":26},"7096dab0-6d27-42d9-b951-7545a5dddf33","why-claude-code-prompt-design-beats-ide-copilots-en","Why Claude Code’s prompt design beats IDE copilots","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778742651754-3kxk.png","2026-05-14T07:10:30.953808+00:00",{"id":76,"slug":77,"title":78,"cover_image":79,"image_url":79,"created_at":80,"category":26},"1f1bff1e-0ebc-4fa7-a078-64dc4b552548","why-databricks-model-serving-is-right-default-en","Why Databricks Model Serving is the right default for production infe…","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778692290314-gopj.png","2026-05-13T17:10:32.167576+00:00",{"id":82,"slug":83,"title":84,"cover_image":85,"image_url":85,"created_at":86,"category":26},"029add1b-4386-4970-bd37-45809d6f7f2f","why-ibm-bob-right-kind-ai-coding-assistant-en","Why IBM’s Bob is the right kind of AI coding assistant","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1778664645900-cyz4.png","2026-05-13T09:30:22.413196+00:00",[88,93,98,103,108,113,118,123,128,133],{"id":89,"slug":90,"title":91,"created_at":92},"8008f1a9-7a00-4bad-88c9-3eedc9c6b4b1","surepath-ai-mcp-policy-controls-en","SurePath AI's New MCP Policy Controls Enhance AI Security","2026-03-26T01:26:52.222015+00:00",{"id":94,"slug":95,"title":96,"created_at":97},"27e39a8f-b65d-4f7b-a875-859e2b210156","mcp-standard-ai-tools-2026-en","MCP Standard in 2026: Integrating AI Tools","2026-03-26T01:27:43.127519+00:00",{"id":99,"slug":100,"title":101,"created_at":102},"165f9a19-c92d-46ba-b3f0-7125f662921d","rag-2026-transforming-enterprise-ai-en","How RAG in 2026 is Transforming Enterprise AI","2026-03-26T01:28:11.485236+00:00",{"id":104,"slug":105,"title":106,"created_at":107},"6a2a8e6e-b956-49d8-be12-cc47bdc132b2","mastering-ai-prompts-2026-guide-en","Mastering AI Prompts: A 2026 Guide for Developers","2026-03-26T01:29:07.835148+00:00",{"id":109,"slug":110,"title":111,"created_at":112},"d6653030-ee6d-4043-898d-d2de0388545b","evolving-world-prompt-engineering-en","The Evolving World of Prompt Engineering","2026-03-26T01:29:42.061205+00:00",{"id":114,"slug":115,"title":116,"created_at":117},"3ab2c67e-4664-4c67-a013-687a2f605814","garry-tan-open-sources-claude-code-toolkit-en","Garry Tan Open-Sources a Claude Code Toolkit","2026-03-26T08:26:20.245934+00:00",{"id":119,"slug":120,"title":121,"created_at":122},"66a7cbf8-7e76-41d4-9bbf-eaca9761bf69","github-ai-projects-to-watch-in-2026-en","20 GitHub AI Projects to Watch in 2026","2026-03-26T08:28:09.752027+00:00",{"id":124,"slug":125,"title":126,"created_at":127},"231306b3-1594-45b2-af81-bb80e41182f2","claude-code-vs-cursor-2026-en","Claude Code vs Cursor in 2026","2026-03-26T13:27:14.177468+00:00",{"id":129,"slug":130,"title":131,"created_at":132},"9f332fda-eace-448a-a292-2283951eee71","practical-github-guide-learning-ml-2026-en","A Practical GitHub Guide to Learning ML in 2026","2026-03-27T01:16:50.125678+00:00",{"id":134,"slug":135,"title":136,"created_at":137},"1b1f637d-0f4d-42bd-974b-07b53829144d","aiml-2026-student-ai-ml-lab-repo-review-en","AIML-2026 Is a Bare-Bones Student Lab Repo","2026-03-27T01:21:51.661231+00:00"]