A library for your research PDFs, on your own machine

What you can do

• Upload PDFs; each is extracted, split into sections and passages, and indexed.
• See each paper’s status while it processes, or the reason it failed.
• Hover a paper to preview its first page and details.
• Fill in title, authors, year, venue and topics from OpenAlex, off until you tick it.
• Correct any detail with Edit details; your corrections survive reprocessing.
• A retracted paper shows a banner at the top of the reader that can’t be dismissed.

How it works

1. Extract

   PyMuPDF pulls out the text with its coordinates, so highlights and citations point at exact spots.

2. Split

   The text is divided into sections and passages sized for search and questions.

3. Index

   Passages are embedded for semantic search once a search model is set up.

Tips

• Start with the ten papers your project rests on, not everything you have.
• Fix a wrong title or year once with Edit details: it stays fixed.

Works well with

• Find papersSearch arXiv, Crossref, CORE, Unpaywall and Semantic Scholar at once.
• WorkspacesGroup papers by project; chat across a whole workspace.
• Semantic searchSearch passages by meaning, with a built-in model or one you pick.
• Local-first and privateOne user, one machine, no account; you choose what leaves it.

Guides

• How to read a research paper efficiently — The three-pass method, what to skip, and how to keep what you read.
• PaperLab vs Zotero — A reference manager and a reading tool do different jobs. Which one you need, or both.
• How to organise research papers — One library, projects instead of folders, notes on the passage.

Terms explained

• Preprint
• Chunking
• Retraction
• Reference manager