SurfSense: Self-Hosted NotebookLM Alternative

A NotebookLM-style research assistant for your own documents: it reads PDFs, Office files, images and audio, answers with citations, and can edit or convert what it finds. It runs as a desktop app or one Docker container, with a local model if you want the work to stay on your hardware.

Pros and cons

Pros

  • Answers cite the pages inside your own documents1
  • Runs air-gapped with a local model2
  • Edits and converts documents, writing a new copy each time3
  • Reads scanned pages through OCR, including Chinese and Japanese4
  • One container, or a desktop app with no server5

Cons

  • Web plugins need a licence after 14 days ($60 first year)6
  • Office conversion needs LibreOffice switched on7
  • Image editing needs an image model that is off by default8
  • Notes cannot link to each other or show a graph9
  • The Docker install keeps your library on the server10

Compare with similar apps

Compare SurfSense with other AI apps. Add or remove apps to see how they differ.

Requirement
SurfSense
PriceFree + PaidFree + Paid
LicenceApache-2.0MIT
Web UIYesYes
Docker Containers1 (app only)None – runs locally
DatabaseSQLiteNone
ArchitectureWorth checking: x86 onlyx86 + ARM
Tech stackPythonNode
DockerYes–
Linux packageYesYes
WindowsYesYes
macOSYesYes
Windows–Yes
macOS–Yes
Linux–Yes
Full comparison

Common questions

Which models can SurfSense use?

Any OpenAI-compatible API or a local model. No model is bundled, so you connect Ollama, vLLM or a hosted provider yourself.

Does SurfSense change my documents?

No. Edits, conversions and the PDF tools always write a new file and leave the original untouched.

Similar apps to SurfSense

Tags

Updated