karakeep vs open-notebook
Self-hostable bookmark-everything app — links, notes, images, PDFs — with AI auto-tagging, OCR, full-text search, full-page archival, RSS auto-hoarding, and a CLI plus skills for LLM agents. — versus — Self-hosted NotebookLM alternative: multi-modal sources, vector plus full-text search, context-aware chat and multi-speaker podcast generation — 18+ model providers incl. Ollama, full REST API.
Two self-hosted personal knowledge hoards with AI on top: Karakeep is capture-first (bookmark everything, auto-tag, archive), Open Notebook is synthesis-first (research a topic, chat over sources, produce podcasts). Collector vs study desk.
| karakeep | open-notebook | |
|---|---|---|
| Stars | 28k | 36k |
| Forks | 1.4k | 4.1k |
| Language | TypeScript | TypeScript |
| License | AGPL-3.0 | MIT |
| Last activity | 2 days ago | yesterday |
| Topics | storage | rag, storage |
| Curated connections | 2 | 2 |
karakeep — the curator's take
The mature choice for self-hosted read-it-later and data hoarding: browser extensions, iOS/Android apps, SSO, importers from Pocket/Omnivore/Linkwarden, monolith full-page archival against link rot, yt-dlp video capture. The AI is seasoning, not the dish — LLM auto-tagging and summarization (works with local Ollama) plus an official CLI and agent skills that turn it into a searchable personal knowledge store OpenClaw/Hermes-class agents can read and write. NOT a RAG framework or team wiki — it's a personal app with collaboration on lists at most. AGPL-3.0, and the stack is real: Meilisearch for search, Puppeteer workers for crawling.
open-notebook — the curator's take
The default answer to 'NotebookLM but private': drop in PDFs, videos, audio and web pages, search and chat across them, and generate podcasts with 1-4 custom speakers where Google caps you at two. Provider freedom is the real lever — 18+ backends down to Ollama/LM Studio keeps sensitive research fully local, and the REST API makes it automatable where NotebookLM is a closed app. NOT ahead on citations — it concedes NotebookLM's source-grounding is stronger (theirs is 'basic, will improve'), which matters if research integrity is the whole point. A product you deploy (Docker), not a library you embed.