Open Notebook

by Luis NovoAppResearch something

Lets you chat with your PDFs, videos and web pages, take notes and turn them into podcasts

View on GitHub

The problem it solves

You have a stack of PDFs, articles and recorded talks for one project and keep scrolling back to find who said what.

For StudentsResearchersAnyone keeping research private

Before: You reread each source, copy quotes into a document by hand and lose track of where each point came from.
After: Your sources sit in one notebook; you ask questions, get answers that cite them, save notes and can turn it all into a podcast.

Stars: ★ 41k
License: MIT
Cost: Usage-based
Platforms: Mac · Win · Linux
Runs: Local · Self-hosted

Screenshots

From the project's own repository.

An Open Notebook notebook with sources on the left, notes in the middle and a chat panel on the right — Image: Luis Novo (MIT). Original file

Overview

Open Notebook is a self-hosted research notebook in the style of Google's NotebookLM that you open in your browser. You add sources such as PDFs, videos, audio, web pages and Office files, then ask questions with answers that cite those sources, write AI-assisted notes, search everything and generate podcasts with one to four speakers. You choose the AI model from more than 20 providers or run local models with Ollama. It suits students, researchers and anyone who wants NotebookLM-style tools on their own computer.

It is a self-hosted NotebookLM-style app where you pick the model provider or run local models, and podcasts can have up to four speakers.

Try these first

Start with the first one: it is a small task you can check easily.

Give me a three-point summary of [source name], and cite where each point comes from.First try · read-only
Compare what [source A] and [source B] say about [topic].
Write a note listing the open questions across all sources in this notebook.saves a note
Make a two-speaker podcast episode from this notebook.uses TTS credits

Example prompts written by Agent Flows Hub.

Features

From the creator's documentation.

Feature	What it does	Runs
Sources	PDFs, videos, audio, web pages, Office documents and pasted text in one notebook	Your computer
Chat with citations	Answers cite the sources, notes and insights they used; you choose what context is shared	Your model provider
Notes	Write notes yourself or have the AI generate insights	Your computer
Search	Full-text and vector search across all your content	Your computer
Podcasts	Episodes with one to four speakers and custom speaker profiles	Your TTS provider
Models	More than 20 providers, or local models through Ollama or LM Studio	Cloud or local

Good to know

Collect the sources for one research topic in a notebook and ask questions that come back with citations. You can then save notes and turn the material into a podcast episode.

Limits

You need Docker and either an API key or a local model server. Add at least one language model and one embedding model; Anthropic and Groq have no embedding models, so pair them with a provider that does.
There is no password by default. If other devices can reach your computer, bind the ports to 127.0.0.1 or set OPEN_NOTEBOOK_PASSWORD before starting.

Notes by Agent Flows Hub (original project name: Open Notebook), written from the creator's documentation, not from running it ourselves. Open Notebook: README and quick start · Open Notebook: Docker Compose guide

Stuck setting it up?

Tell us what happened. We take a look first; if it needs paid work, you get an estimate before anything starts.

Get help

How to start

Install itPaste the install request into an AI agent you already use, or run the official command yourself or follow the official guide.macOS, Windows, Linuxmkdir open-notebook && cd open-notebook curl -o docker-compose.yml https://raw.githubusercontent.com/lfnovo/open-notebook/main/docker-compose.yml
Open itFollow the first-run steps in the app.
Give it a first taskCreate a notebook, add one web page you know well as a source, and ask for a three-point summary. Check that each point cites the source.

Steps follow the creator's instructions. The request tells your agent to read them first and wait for your OK.

Before you start

A computer with macOS, Windows or Linux.
Docker Desktop on macOS or Windows, or Docker Engine with Compose on Linux
An API key from an AI provider, or a local model server such as Ollama
Your own encryption key in docker-compose.yml, then docker compose up -d
Pay per use. The AI service it calls charges for what you use.

Details

Made by: Luis Novo
License: MIT Terms
Platforms: macOS, Windows, Linux
Runs on: Your computer, Self-hosted server

View source Official docs Report a problem

For AI agents

This page as a Markdown install brief:

/en/library/open-notebook/install.md

Other apps

Open WebUI Gives you and your team a private ChatGPT-style workspace on your own computer or server
OpenChatCut Lets you edit videos by chatting with an AI, then fine-tune the cuts on a real timeline
OpenCode Telegram Bot Lets you send coding tasks to OpenCode on your computer from Telegram on your phone
OpenCreator Translates and dubs videos, writes scripts and posts, and makes images and videos
OpenDesign Turns a short brief into web prototypes, slide decks, images and motion graphics
OpenFilm Lets your coding agent make animated videos that you can then change by hand in an editor

See all 113 apps

GitHub