RAG Open-Source Projects

Showing 40 indexable projects from the current snapshot.

• open-webui/open-webui

  A self-hostable web interface for managing and interacting with LLM runners and AI features. It is designed to operate entirely offline and supports various model providers alongside built-in retrieval augmented generation.

• langgenius/dify

  Dify is an open-source platform for teams building, testing, and deploying agentic AI workflows and LLM applications through a visual interface, RAG features, agent tools, model management, and observability. It can be self-hosted or used…

• langchain-ai/langchain

  LangChain is a Python framework that simplifies building LLM applications by connecting models, embeddings, and vector stores through interoperable components. Developers can chain these integrations together using flexible abstractions ra…

• infiniflow/ragflow

  RAGFlow provides a retrieval-augmented generation workflow focused on deep document understanding, processing complex unstructured formats to extract knowledge with traceable citations. It requires external LLM and embedding services to op…

• Shubhamsaboo/awesome-llm-apps

  A repository of over 100 open-source AI agent and RAG application templates, organized across multiple domains such as voice, generative UI, and coding tools. It serves as a reference for developers and general users seeking tested startin…

• run-llama/llama_index

  A data framework for augmenting LLMs with private data and building agentic applications, offering data connectors, indexing, and retrieval/query interfaces. It is designed for development teams that need to ingest existing documents, APIs…

• unclecode/crawl4ai

  Crawl4ai converts web pages into clean, LLM-ready Markdown and structured JSON for retrieval pipelines and AI agents. It provides controllable browser-based crawling and data extraction without mandatory API keys, operating as a self-hoste…

• microsoft/generative-ai-for-beginners

  A 21-lesson course teaching how to build generative AI applications using Python and TypeScript. It covers core concepts and provides code examples, requiring learners to have basic programming knowledge and to choose an LLM provider to ru…

• Mintplex-Labs/anything-llm

  This application lets users chat with their documents and run AI agents in a private, locally hosted environment by default. It supports multi-user permissioning, dynamic model routing, and custom agent construction without requiring codin…

• Graphify-Labs/graphify

  Graphify converts local project directories into queryable knowledge graphs using deterministic tree-sitter AST parsing across approximately 40 programming languages without requiring LLM calls. It outputs an interactive HTML visualization…

• milvus-io/milvus

  Milvus is a cloud-native vector database for organizing and searching unstructured data at scale. It supports vector ANN search, hybrid search with metadata filtering, and streaming updates for AI applications.

• FlowiseAI/Flowise

  Flowise provides a visual, drag-and-drop canvas for assembling and deploying AI agents and multi-agent workflows without writing code. It runs locally on Node.js or on multiple cloud hosting platforms and is intended for users who need to…

• PaddlePaddle/PaddleOCR

  PaddleOCR converts PDF documents, images, and Office files into structured formats including JSON, Markdown, and DOCX. The project features PaddleOCR-VL-1.6, a 0.9B parameter vision-language model, alongside PP-StructureV3 layout conversio…

• weaviate/weaviate

  Weaviate is a cloud-native vector database that unifies vector similarity search, hybrid keyword filtering, and retrieval-augmented generation (RAG) into a single query. It is designed for developers and AI engineers building scalable AI a…

• Egonex-AI/Understand-Anything

  Understand Anything converts source code repositories and LLM wiki directories into an interactive, explorable knowledge graph dashboard. It uses a hybrid Tree-sitter and LLM pipeline to map structural architecture and business logic, and…

• opendatalab/MinerU

  MinerU converts PDF, image, DOCX, PPTX, and XLSX files into Markdown and JSON for retrieval and extraction workflows. It runs locally across multiple backends via CLI or WebUI and requires at least 16GB of RAM, with optional GPU accelerati…

• microsoft/ai-agents-for-beginners

  An 18-lesson course that teaches the fundamentals of building AI agents through written guides, videos, and Python code samples. It covers core topics from design patterns to production deployment using Microsoft Agent Framework.

• datawhalechina/hello-agents

  Adopt this as a Chinese-language tutorial when your goal is to learn how to build AI Native Agents through theory, hands-on agent paradigms, and multi-agent examples. It is a learning resource rather than documented production deployment g…

• assafelovic/gpt-researcher

  An autonomous research agent that aggregates over 20 web or local sources in parallel to produce detailed, cited reports. It requires external LLM and search API providers and is positioned as an experimental application provided without w…

• ruvnet/ruflo

  Ruvnet/ruflo adds an execution and coordination layer around Claude Code and OpenAI Codex, providing specialized agents, coordinated swarms, self-learning memory, and zero-trust federation. It targets developers and enterprise teams who ne…

• docling-project/docling

  docling-project/docling parses diverse document formats—including PDF, DOCX, PPTX, XLSX, HTML, images, and audio—and prepares them for generative AI applications. It provides advanced PDF understanding, a unified document representation fo…

• headroomlabs-ai/headroom

  This project reduces LLM token consumption from agent contexts through local-first, reversible context compression. It caches compressed originals locally so the LLM can retrieve them on demand, lowering costs without losing accuracy.

• pathwaycom/pathway

  A Python framework with a Rust engine for building scalable ETL, streaming, real-time analytics, and RAG pipelines. It processes batch and streaming data using the same Python code.

• mem0ai/mem0

  Mem0 provides an intelligent memory layer for AI assistants and agents, retaining user, session, and agent state for adaptive personalization. It uses an external LLM by default to process conversational inputs and returns stored or retrie…

• pathwaycom/llm-app

  Ready-to-deploy templates for production RAG and enterprise search that natively synchronize live data sources. Built-in indexing handles vector, hybrid, and full-text search without requiring external vector databases.

• MemPalace/mempalace

  mempalace/mempalace is a local-first AI memory application that stores conversation history verbatim, preventing context loss during AI coding sessions. It provides scoped semantic search, pluggable storage backends, and 36 MCP server tool…

• zylon-ai/private-gpt

  zylon-ai/private-gpt provides an API-first backend layer that follows the Claude API model to build applications on local models without rebuilding backend primitives or relying on cloud APIs. It requires a separately hosted OpenAI-compati…

• stanford-oval/storm

  STORM automates the research and writing process to generate long, Wikipedia-style articles with citations from scratch. It uses multi-perspective question asking and simulated conversations with topic experts grounded in Internet sources…

• anthropics/claude-cookbooks

  A collection of Jupyter notebooks and recipes providing code examples for building applications with the Claude API. It covers classification, retrieval augmented generation, summarization, tool use, and multimodal tasks.

• abhigyanpatwari/GitNexus

  GitNexus indexes Git repositories or ZIP files into a precomputed knowledge graph using Tree-sitter, exposing structural context through 17 MCP tools or a browser-based UI. It supports AI coding agents by providing dependencies, call chain…

• QuivrHQ/quivr

  quivr-core is an opinionated Python library that helps developers build Retrieval-Augmented Generation (RAG) systems quickly, supporting multiple LLM providers and file types. Adopt this project to reduce RAG design overhead and integrate…

• chatchat-space/Langchain-Chatchat

  Langchain-Chatchat is a locally deployable RAG and Agent application designed for Chinese-language scenarios and open-source models. It requires a separate model inference framework to provide offline knowledge base Q&A, agent tool calling…

• deepseek-ai/awesome-deepseek-integration

  A curated, categorized directory maintained by the deepseek-ai organization for discovering software, tools, and frameworks that integrate with the DeepSeek API. It serves as a reference hub for agent frameworks, RAG tools, and editor exte…

• HKUDS/LightRAG

  LightRAG is a graph-based retrieval-augmented generation framework that reduces computational overhead and LLM calls during indexing and querying compared to traditional GraphRAG. It supports incremental knowledge base updates without glob…

• google/langextract

  This Python library extracts structured information from unstructured text documents such as clinical notes and reports. It provides source grounding for every extraction, processes large documents using chunking and parallel processing, a…

• patchy631/ai-engineering-hub

  A collection of 93+ production-ready projects and tutorials spanning beginner to advanced AI engineering topics, including large language models, retrieval-augmented generation, and agents. It functions as a structured learning repository…

• lfnovo/open-notebook

  A self-hosted AI research notebook enabling data sovereignty over multi-modal research materials. It provides content ingestion, context-aware chat, search, multi-speaker podcast generation, and configurable AI model routing.

• khoj-ai/khoj

  Khoj is a self-hostable AI assistant that lets users privately search and query their own documents and the web using local or cloud-based LLMs. It supports text, voice, and image inputs, and is accessible via web, desktop, mobile, and bro…

• ItzCrazyKns/Vane

  A self-hosted AI answering engine that runs on your own hardware to keep searches private. It combines local or cloud LLMs with web search to deliver cited answers.

• microsoft/graphrag

  GraphRAG uses LLMs to extract structured data from unstructured text and applies knowledge graph memory structures to enhance LLM reasoning over private data. It is a Python library and demonstration project that requires CLI usage, prompt…