RAG・知識ベース
insights-lm-public
Self-hosted NotebookLM alternative — chat with documents, get audio summaries.
★ 656 スターMIT
ガイドを見る →SurfSenseは、MODSetter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: MODSetter/SurfSense）。GitHubスターは16,335、フォークは1,569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10月7日です。
Air-gapped, open-source NotebookLM alternative that turns local documents into briefings, decks and podcasts.
SurfSense fits when files are too sensitive to upload to a hosted notebook tool and you still want cited answers and finished artifacts from them.
Get the signed installer for Windows, macOS or Linux from the project's releases or download page; no Docker, terminal or account is required.
Let the app pull a local model for you, or paste a base URL and key for any OpenAI-compatible API; keys are stored encrypted in your OS keychain.
Drop in PDFs, Office files or images, ask cited questions, then open Studio to generate a briefing, deck, study guide or podcast from the same sources.
Parsing, chunking and embedding happen on your computer into a local SQLite index; the app uploads nothing and has no telemetry.
Ask questions across all your documents and every answer cites the source it came from.
Turn sources into summaries, flashcards, quizzes, mind maps, editable .pptx slides, .docx reports, .xlsx spreadsheets, HTML pages and PDFs.
Generate a two-host audio conversation from your documents, voiced offline by the bundled Kokoro-82M model.
Use a local model the app downloads for you, or point SurfSense at any OpenAI-compatible endpoint, including your own local server.
Outbound connections are off by default, with an egress panel listing every destination the app can reach so you choose what to enable.
Clone the repo if you want the open-source Docker stack or to work on the code; most users simply download the desktop installer instead.
git clone https://github.com/MODSetter/SurfSense.git
cd SurfSenseInstall the signed desktop build for your platform from the releases page, which is the supported path in the README.
# Download SurfSense-Setup.exe / SurfSense-arm64.dmg / SurfSense.AppImage from GitHub Releases - see README for the exact linkOpen the app, let it pull a local model or add an OpenAI-compatible endpoint, then drop in your first documents.
# Launch SurfSense and follow the in-app model picker - see README for the exact commandIt is an open-source project by MODSetter and community contributors, with a Discord and GitHub Discussions for support and direction.
The app and its updates are free and the code is Apache-2.0 licensed. The README notes an optional licence that adds plugins and priority support but gates nothing else.
No. Every role can run locally: you can let the app download a local model and work with no account or API key, or optionally bring your own OpenAI-compatible key.
Both produce summaries, flashcards, quizzes, mind maps and audio overviews, but SurfSense runs offline on your machine with your choice of model, while NotebookLM uploads sources to Google and uses Gemini only.
Download the installer for your platform, pick a model in the app, drop in documents, and ask your first cited question - no terminal or Docker needed.
Self-hosted NotebookLM alternative — chat with documents, get audio summaries.
Knowhere turns documents into structured, hierarchy-native chunks ready for agents and RAG.
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms
Skills that let coding agents search and read your local documents.
OpenKB compiles your documents into a persistent, interlinked wiki knowledge base — no vector DB
A local-first AI knowledge base and NotebookLM replacement built with Electron.
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 — SurfSense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SurfSense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。
A blazing-fast vector index in Rust, built on TurboQuant - with Python bindings
Lightweight, lightning-fast in-process vector database from Alibaba that embeds in your app.
A private AI platform for agents, assistants and enterprise search over your documents
Unified framework for enterprise RAG pipelines with small, specialized models - local, private and secure.
An agentic AI data assistant - chat with your databases, run SQL and code, get reports
Local knowledge-base Q&A over PDFs, Office files, images and web pages - offline and CPU-friendly.