PaperQA (paper-qa)
Project record
Maintained by FutureHouse134
PaperQA is a Python package and command-line tool (pqa) from FutureHouse for retrieval-augmented question answering over PDFs, text, Office documents, and source code, focused on scientific literature. Versions 5 and later, called PaperQA2, add an agent that searches papers, gathers evidence, and writes answers with in-text citations; the package moved to calendar versioning in December 2025.2
- Repository: GitHub repository (external site: github.com)
- License: LICENSE (Apache 2.0) (external site: github.com)
- Release notes: PyPI package (external site: pypi.org)
- Documentation: Documentation (hosted by Edison Scientific) (external site: docs.edisonscientific.com)
- Paper: PaperQA2 paper (arXiv 2409.13740) (external site: arxiv.org)
Availability and license
Overall availability
Source code is public and the package installs from PyPI. Running it requires access to an LLM and embedding model (an API key for a hosted provider, or a local model server); indexing many papers may also call for Crossref and Semantic Scholar API keys. Users supply their own PDFs.24
Availability is separate from permission: read the license before using or redistributing.
The LICENSE file is Apache 2.0 with a 2024 FutureHouse copyright notice. It covers this package only; the LLM providers, metadata services, and papers it works with have their own terms. The README notes that FutureHouse's internal version uses paper-access licenses it cannot share, so published results may differ from what the open package produces.32
Component reuse rights
- weights
- Unknown — no complete fact-level rights review
- code
- Reviewed qualifying license recorded — check scope and conditions
- data
- Unknown — no complete fact-level rights review
- documentation
- Unknown — no complete fact-level rights review
No complete system-rights review is recorded for this release.
Public materials checklist
| Item | Status | Notes and evidence |
|---|---|---|
| Source codeIs the source code publicly readable? | Public | Public GitHub repository under the Future-House organization.1 |
| DocumentationIs user documentation published? | Public | The README covers installation, CLI and library use, model and embedding choices, settings, and reproduction; Edison Scientific's documentation site also hosts PaperQA2 documentation.25 |
| InstallationAre installation instructions or packages publicly available? | Public | Installed with pip from PyPI; optional extras add local embedding models.24 |
| Supported platformsAre supported operating systems or hardware documented? | Partial | PyPI metadata requires Python 3.11 or later and lists the package as OS independent. No hardware or operating-system support matrix was found in the pages reviewed.4 |
| Release statusAre versioned releases published? | Public | Versioned releases are published on PyPI; the latest at review was 2026.8.12, uploaded on August 12, 2026.4 |
What it is useful for
Answering questions over a local collection of research papers with cited answers, building a full-text index of papers, and fetching paper metadata such as citation counts and retraction checks. It works with any LiteLLM-compatible model, including locally hosted models.2
Run and use notes
- By default PaperQA2 uses OpenAI embeddings and models with a NumPy vector store; the README documents switching to other LiteLLM providers, locally hosted models, and Sentence Transformers embeddings.2
- The README lists the LitQA2 question IDs and paper DOIs used for its train, evaluation, and test splits so published PaperQA2 evaluations can be reproduced.2
- Edison Scientific's PaperQA3, announced in February 2026, runs in Edison's hosted agents and is not part of this open-source package; repository maintainers stated in July 2026 that there were no plans to open-source it.67
Organization context
U.S. eligibility
Eligible · basis: U.S.-governed project
The repository is published under FutureHouse's GitHub organization, the LICENSE names FutureHouse as copyright holder, and the PyPI package lists FutureHouse technical staff as author. FutureHouse is a 501(c)(3) nonprofit in San Francisco, per its about page and the IRS exempt-organization extract. Edison Scientific, a spinout of FutureHouse, hosts PaperQA2 documentation; the sources reviewed do not say that maintenance moved to Edison.134895
Sources
This listing is not an endorsement, a safety assessment, or a federal approval.