Documents and knowledge · Independent · 21 Oct 2026

Future-House/paper-qa

High accuracy RAG for answering questions from scientific documents with citations

← Documents and knowledge
PaperQA2: scientific document RAG. Answers questions from research papers with exact citations.

In simple English

This tool reads scientific research papers and answers questions about them. It extracts text, fetches citation details, and builds a searchable index from local PDF files. It uses artificial intelligence to find the right passages and write answers with exact citations. You need an API key from an artificial intelligence provider or a local server to run it. It saves time for researchers who need to search through many documents at once.

What it is

PaperQA2 is a Python library and command-line tool for agentic retrieval augmented generation (a way to let a language model search local files and synthesize grounded answers). It parses documents, retrieves relevant chunks using vector embeddings and keyword search, and generates responses with in-text citations.

What it is good for

Solo builders and researchers who need to query a large collection of scientific papers, technical documents, or source code files and get verified answers with source references.

Stars9.4k
LicenceApache-2.0
Latestv2026.08.12 (2026-08-12)checked 21 Oct 2026
Last code change2026-10-20
Written inPython
Setup effort
3 / 5
Payoff
4 / 5
Good
  • Automatically fetches paper metadata, citations, and retraction checks from services like Crossref and Semantic Scholar.
  • Supports multiple document formats including PDF, text, Microsoft Office files, HTML, and source code.
  • Works with many language model providers through LiteLLM, plus local models.
Watch for
  • Requires Python 3.11 or newer and familiarity with the command line.
  • Consumes significant numbers of API tokens when indexing large sets of research papers.
  • Index files built with older versions are incompatible and must be rebuilt.
Replaces

Manual literature reviews and reading through hundreds of PDF files by hand.

Works well with

LiteLLM or OpenAI API for language model access.

Similar repositories
  • huggingface/sentence-transformers19k stars

    Provides local embedding models that can be used with PaperQA2 for offline vector search.

  • run-llama/llama_index53k stars

    A larger data framework for connecting custom documents to language models, whereas PaperQA2 focuses specifically on scientific papers and citations.

Also reviewed on Repo Radar
  • Hack-with-Github/Awesome-Hacking123k stars

    An index of 81 curated security reading lists, with one line on each.

  • Zettlr/Zettlr14k stars

    A desktop writing program for Markdown documents, with citation support and export to Word, PDF and LaTeX.

  • JabRef/jabref4.8k stars

    A desktop citation and reference manager that stores its library as a plain BibTeX file.

Install
pip install paper-qa>=5

Independent: reviewed for Repo Radar outside the daily editions. Every independent repository

On Stack Radar: AI & Automation › AI Assistants & Agents › Research Agents.

Get the next edition in your inbox

A dozen repositories, opened and checked. The licence read, the last release dated, and the ones that did not make it named with the reason. It is the half most lists leave out.

No tracking pixels. One click to leave. The archive stays free either way.

We use your address to send the edition and nothing else. Confirm by email, leave in one click. How we handle it.