RAG・知識ベース
ezdata
AI-native data platform: 60+ connectors, ETL, scheduling, and RAG search.
★ 293 スターMIT
ガイドを見る →swirl-searchは、swirlaiが公開しているオープンソースのRAG・知識ベースプロジェクトです（GitHub: swirlai/swirl-search）。GitHubスターは3,047、フォークは288で、Apache-2.0ライセンスの下で公開されています。GitHubでの最終更新は2026年9月26日です。
Federated AI search and RAG across 100+ apps — answers with sources, without moving your data
List the sources you want to search in one place — the README notes connectors across 100+ apps, and it searches Arxiv, European PMC and Google News out of the box.
Download the project's docker-compose.yaml (or clone the repo) with Docker installed and running.
Optionally set your own OpenAI key and host variables for real-time RAG, then pull and start the containers.
Open the Galaxy UI, log in, and run a question across your connected sources to see ranked, cited answers.
Queries run live against your sources in place, with the user's own permissions enforced at the source.
Optionally generates an answer using the LLM of your choice, with sources you can click through to verify.
No vector DB, ETL pipeline or duplicated copy of your data to secure and audit.
Connects across a large library of enterprise and public sources (see the project's connector list).
Starts with one Docker Compose flow and is usable in about two minutes per the README.
The open-source Community edition is Apache-2.0; a separate Enterprise edition adds features such as a three-pass reranker and an MCP server.
Clone SWIRL Search and enter the directory (alternatively just download its docker-compose.yaml).
git clone https://github.com/swirlai/swirl-search.git
cd swirl-searchAs the README quick start shows, you can fetch the compose file directly.
curl https://raw.githubusercontent.com/swirlai/swirl-search/main/docker-compose.yaml -o docker-compose.yamlPull and start the containers, then open http://localhost:8000/galaxy and log in with admin / password.
docker-compose pull && docker-compose upIt is developed by swirlai (SWIRL AI) as the open-source SWIRL Community edition, alongside a commercial Enterprise edition.
The Community edition in this repo is free and Apache-2.0 licensed. Enterprise features and managed support are a separate commercial offering.
Federated search works without one; real-time RAG with OpenAI requires your own OpenAI API key, set via the OPENAI_API_KEY environment variable.
Typical stacks copy and embed all your data first. SWIRL queries sources live, in place, so there is no duplicate corpus to govern.
Install Docker, download docker-compose.yaml, run docker-compose pull && docker-compose up, then open http://localhost:8000/galaxy.
AI-native data platform: 60+ connectors, ETL, scheduling, and RAG search.
Self-hosted RAG search across GitHub, Drive, Notion, and local files.
Ready-to-run templates for RAG, AI pipelines and enterprise search that stay in sync with live data
An agent that finds anything on your computer — federated search over your files, in place.
SQLBot — DataEase's conversational analytics: ask in natural language, get SQL answers via LLMs plus RAG.
Free rag & knowledge tool that 🥥 coco ai server - search, connect, collaborate, ai-pow.
LibreAIカタログに掲載されたRAG・知識ベースプロジェクトの中で最もスターの多いもの — swirl-searchが合わなかった場合に検討したい、現実的な候補リストです。
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
カタログの中でswirl-searchとスター数が近いRAG・知識ベースプロジェクト — 前後の棚もそのまま見て回れます。
seekdb — the AI-native search database unifying vector, text and structured data for agents.
Graph RAG conversational knowledge base built on TiDB Serverless vector storage
Fast, lightweight Python embeddings from Qdrant — dense, sparse, ColBERT, image and rerankers.
The semantic intelligence layer for ontologies — governed, traceable knowledge for AI agents
VideoRAG — chat with your videos, even hundreds of hours of them, with graph-driven retrieval.
All-in-one API platform for search, recommendations, RAG and analytics — self-hostable