HomeProjectsArticlesPK BoardAbout
Explore Projects

RAGFlow

Active
GitHub Go Apache-2.0

Description

A leading open-source RAG engine that fuses cutting-edge retrieval-augmented generation with agent capabilities to create a superior context layer for LLMs.

Key Features

  • Deep document understanding - Extracts knowledge from complex unstructured data via deepdoc, supporting PDF, Word, PPT, Excel, images, and more
  • Template-based chunking - Multiple intelligent chunking templates for explainable document segmentation and knowledge organization
  • Traceable citations with reduced hallucinations - Visualizes text chunks for human intervention, answers include key reference sources
  • Agent workflow and MCP - Supports orchestrable agent workflows with MCP protocol and built-in Python/JavaScript code executor
  • Heterogeneous data source compatibility - Handles Word, PPT, Excel, images, scanned documents, structured data, and web pages
  • Cross-language query and data sync - Supports cross-language retrieval with sync from Confluence, S3, Notion, Discord, and Google Drive

Use Cases

💡 Enterprise knowledge base construction, transforming internal documents, contracts, and reports into searchable knowledge stores
💡 Intelligent customer service, providing precise context to LLMs from enterprise docs to reduce answer hallucinations
💡 Legal and compliance document analysis, quickly retrieving and citing key clauses from large volumes of legal files
💡 Multi-modal document Q&A, parsing PDFs/DOCXs with embedded images for cross-language intelligent问答
💡 R&D knowledge management, syncing technical docs and codebases for unified team knowledge retrieval

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (88.6k stars)
  • • Permissive open-source license (Apache-2.0)
  • • Established track record (2 years in production)

Tags

rag document-understanding knowledge-base retrieval ocr python

Categories

📚 RAG Tools

Quick Start

# Ensure Docker >= 24.0.0 and Docker Compose >= v2.26.1
# Set system parameter
sysctl -w vm.max_map_count=262144

# Clone and start
git clone https://github.com/infiniflow/ragflow.git
cd ragflow
docker compose up -d

# Access web UI
# Open http://localhost:80 in browser
# Configure LLM and embedding models in UI, upload documents to start RAG Q&A
Visit GitHub Visit Website

Related Projects

LightRAG

38.9k · Python
Active A+

LightRAG is a simple and fast Retrieval-Augmented Generation framework using graph-enhanced retrieval, published at EMNLP 2025.

raggraphretrieval +2
  • · Graph-enhanced retrieval — Dual-level retrieval (local/global) using knowledge graph entities and relationships, more precise than vector search
  • · Four text chunking strategies — Fixed, Recursive, Vector, and Paragraph chunking adapted to different document types
  • · Multi-backend storage — Neo4j, PostgreSQL, MongoDB, OpenSearch, JSON KV Store and other storage backends supported
View full details → View on GitHub

RAG-Anything

22.9k · Python
Active A+

All-in-one RAG framework supporting text, images, tables, equations and more document formats for retrieval-augmented generation with unified knowledge QA.

ragmultimodalknowledge-base +2
  • · End-to-end multimodal pipeline — Complete workflow from document ingestion to intelligent multimodal query answering
  • · Universal document support — Seamless processing of PDFs, Office documents, images and diverse file formats
  • · Specialized content analysis — Dedicated processors for images, tables, mathematical equations and heterogeneous content
View full details → View on GitHub

Haystack

26.2k · Python
Active A+

Haystack is an enterprise-grade framework for RAG and search applications, covering document processing, retrieval, generation, and evaluation end to end.

ragretrievalllm +1
  • · Modular pipeline architecture — compose retrieval, routing, memory, and generation components as DAGs with conditional branches and loops
  • · Model and vendor agnostic — swap OpenAI, Anthropic, Cohere, Hugging Face, AWS Bedrock, and local models without rewriting
  • · Context engineering control — explicit control over how information is retrieved, ranked, filtered, combined, and routed before reaching the model
View full details → View on GitHub