// rag · VectifyAI/OpenKB

OpenKB

OpenKB est un projet de RAG et connaissances open source, créé par VectifyAI (VectifyAI/OpenKB sur GitHub). Il compte 4 759 étoiles sur GitHub et 495 forks, il est publié sous la licence Apache-2.0 et a été mis à jour pour la dernière fois sur GitHub le 22 juillet 2026.

OpenKB compiles your documents into a persistent, interlinked wiki knowledge base — no vector DB

★ 4,8 k étoiles⑂ 495 forksApache-2.0mis à jour juil. 2026

Voir sur GitHub ↗Official site ↗

▦ visuel — le README de ce projet ne contient aucune capture intégrable,
voici donc un emplacement réservé. Le guide ci-dessous est complet malgré tout.

// détails du projet

Détails du projet

Dépôt: VectifyAI/OpenKB
Étoiles GitHub: 4 759
Forks: 495
Licence: Apache-2.0
Dernière mise à jour sur GitHub: 22 juillet 2026
Catégorie: RAG et connaissances
Type de fiche: Guide d'installation complet
Site officiel: Official site

// fonctionnement

Comment ça marche

Gather your documents
Collect the papers, docs, or URLs you want compiled into one knowledge base.
Install and initialize
Install OpenKB from PyPI, create a folder, and run the init command, choosing your LLM model.
Add documents
Add files, a whole directory, or a URL; OpenKB compiles them into wiki pages with cross-links.
Query and chat
Ask questions with cited answers, chat interactively, or generate outputs like skills and slide decks.

// fonctionnalités

Pourquoi on l'adore

Compiled wiki, not chunk RAG

Documents become summary, concept, and entity pages with cross-references that accumulate over time; contradictions get flagged.

Vectorless long-document retrieval

PageIndex tree indexing handles long PDFs by reasoning over a hierarchical index — no vector database required.

Broad format support

PDF, Word, Markdown, PowerPoint, HTML, Excel, CSV, text, and URLs, with native handling of figures, tables, and images.

Query and chat with sessions

One-off cited answers or multi-turn chat over the wiki, with persisted sessions you can resume.

Skill Factory and generators

Distill redistributable agent skills from the wiki, visualize an interactive knowledge graph, or generate single-file HTML slide decks.

Obsidian-compatible and OKF-ready

The wiki is plain Markdown with wikilinks that opens in Obsidian, and pages follow Google's Open Knowledge Format specification.

// lancez-le

Lancez-le vous-même

Clone the repository
Clone for a source install, or simply install from PyPI in the next step.
```
git clone https://github.com/VectifyAI/OpenKB.git
cd OpenKB
```
Install OpenKB
Install the package (or `pip install -e .` from the clone for development).
```
pip install openkb
```
Initialize and query
Create a knowledge base, add a document, and ask your first question (set your LLM API key in .env first).
```
openkb init
openkb add paper.pdf
openkb query "What are the main findings?"
```

// cas d'usage

Ce que vous pouvez créer

Turn a folder of research papers into a searchable, cross-linked wiki
Maintain a team knowledge base that stays current as new documents are added
Ask cited questions over very long PDFs that exceed model context windows
Distill a book's worth of material into a portable agent skill

// faq

Questions fréquentes

Who maintains OpenKB?

VectifyAI, the team behind PageIndex; OpenKB is powered by PageIndex's open-source vectorless retrieval.

Is OpenKB free?

Yes, it is Apache-2.0 licensed. You pay only for the LLM API usage of the provider you configure (OpenAI, Claude, Gemini, and others via LiteLLM); some subscription providers can authenticate via OAuth instead of a key.

Does it need a paid API key?

It needs an LLM — usually your own API key set in .env. PageIndex runs locally by default; the optional PageIndex Cloud features need a separate key.

How does it compare to traditional RAG tools?

RAG rediscovers knowledge from chunks on every query and nothing accumulates. OpenKB compiles knowledge once into a persistent wiki with cross-references already built, then keeps it current as documents are added.

How do I get started?

Les autres projets de la catégorie RAG et connaissances avec le plus d'étoiles du catalogue LibreAI — la sélection réaliste d'alternatives si OpenKB ne vous convient pas.

MarkItDown★ 193,5 k
MarkItDown — Microsoft tool that converts PDFs, Office files, images and more to clean Markdown for LLMs, MIT
Dify★ 157,8 k
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
ragflow★ 91,6 k
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context…
PaddleOCR★ 90,8 k
Lightweight OCR toolkit turning PDFs and images into structured, LLM-ready data in 100+ languages
MinerU★ 82,5 k
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms

// plus en rag

Plus en RAG et connaissances

Projets de la catégorie RAG et connaissances proches dans le catalogue, les plus proches en étoiles de OpenKB — continuez à explorer de part et d'autre.

AutoRAG★ 5,1 k
An agent that finds anything on your computer — federated search over your files, in place.
infinity★ 4,7 k
AI-native database for hybrid search across dense, sparse, and tensor vectors plus full text
code-graph-rag★ 5,2 k
RAG for your monorepo: query, understand and edit multi-language codebases with a knowledge graph.
ragapp★ 4,4 k
RAGApp: deploy a RAG app over your documents with Docker — configure it like a custom GPT
llm-graph-builder★ 5,3 k
Build Neo4j knowledge graphs from unstructured data using LLMs — then chat with your graph.
ragent★ 4,2 k
Ragent (容纳智能体): a production-grade Java Agentic RAG platform with hybrid retrieval and an admin console

Obtenir OpenKB

Gratuit et ouvert : récupérez le code source, mettez une étoile au dépôt et commencez à créer.

Voir sur GitHub ↗← Plus d'outils RAG et connaissances