RAG et connaissances
llm-graph-builder
Build Neo4j knowledge graphs from unstructured data using LLMs — then chat with your graph.
★ 5,3 k étoilesApache-2.0
Voir le guide →graphrag est un projet de RAG et connaissances open source, créé par microsoft (microsoft/graphrag sur GitHub). Il compte 36 264 étoiles sur GitHub et 3 835 forks, il est publié sous la licence MIT et a été mis à jour pour la dernière fois sur GitHub le 8 octobre 2026.
Microsoft's graph-based RAG: index a corpus into a knowledge graph, then query it with an LLM
Decide what you want to achieve with graphrag and check the README on GitHub for requirements before installing.
Clone the repository from GitHub (or take the packaged install route in step 3 if the README offers one).
Follow the project's documented install commands — the Run section below lists the exact ones — and add any required API keys or models.
Work through the README's quickstart example end to end, then adapt it to your own project or data.
An LLM extracts entities, relationships and claims from your documents into a structured graph during indexing.
Query modes cover both corpus-wide sensemaking questions (global search over graph communities) and specific entity questions (local search).
Includes tooling to tune the extraction and summarization prompts to your domain for better index quality.
The indexing pipeline is configurable, so you can adapt chunking, extraction and embedding steps to your data.
A maintained docs site walks through the CLI quickstart, configuration and responsible-AI considerations.
Published by Microsoft Research as a methodology for enhancing LLM outputs with knowledge-graph memory structures.
Get the GraphRAG source and documentation.
git clone https://github.com/microsoft/graphrag.git
cd graphragInstall GraphRAG from PyPI into a fresh Python environment.
pip install graphragCreate a project, drop documents into its input folder, add your LLM key to the generated .env, then run indexing and a first query.
graphrag init --root ./ragtest
# add your documents to ./ragtest/input and your API key to ./ragtest/.env, then:
# graphrag index --root ./ragtest (see the docs quickstart)graphrag is maintained by the microsoft organization/account on GitHub, together with open-source contributors, in the public repository microsoft/graphrag.
Yes — the code is open source under the MIT license. You can use, modify and redistribute it under that license's terms.
Yes — GraphRAG calls an LLM (and usually an embedding model) during indexing and querying, so you need a paid provider API key, and Microsoft warns indexing large corpora can be expensive; start small.
Vector RAG is cheaper and simpler for lookup questions; GraphRAG's graph index handles global, sensemaking questions better but costs more to build and maintain.
Clone the repository from GitHub, follow the install commands in the Run section (taken from the project's README), and complete the documented quickstart example before adapting it to your own use case.
Build Neo4j knowledge graphs from unstructured data using LLMs — then chat with your graph.
RAG for your monorepo: query, understand and edit multi-language codebases with a knowledge graph.
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms
Graph RAG conversational knowledge base built on TiDB Serverless vector storage
Self-hosted rag & knowledge: turn your own pdf textbooks into an ai-searchable knowl.
Les autres projets de la catégorie RAG et connaissances avec le plus d'étoiles du catalogue LibreAI — la sélection réaliste d'alternatives si graphrag ne vous convient pas.
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context…
Lightweight OCR toolkit turning PDFs and images into structured, LLM-ready data in 100+ languages
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms
Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience
Projets de la catégorie RAG et connaissances proches dans le catalogue, les plus proches en étoiles de graphrag — continuez à explorer de part et d'autre.
Local knowledge-base RAG and agent app built on LangChain with ChatGLM, Qwen and Llama models
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
PageIndex: vectorless, reasoning-based RAG that indexes documents like a human expert would
A knowledge-base platform with visual AI workflows — build Q&A systems on your own documents
Open-source engine turning continuous content streams into search, monitoring and alerts
Turn PDFs into AI-ready data: extract Markdown, JSON and tables — and auto-tag for accessibility
Gratuit et ouvert : récupérez le code source, mettez une étoile au dépôt et commencez à créer.