When you select a paper in Papertree, the details panel shows an availability section with colored badges, version cards, and access links. This page explains what all of that means.
Every paper is classified into one of six access categories. These come from OpenAlex, which tracks where papers can be read for free across the internet. The categories follow a priority system: if a paper is free at the publisher, that takes precedence over a free copy in a repository.
Published in a fully open-access journal that charges no fees to authors or readers. These journals are funded through other means (institutions, grants, societies). This is the most "open" a paper can be — free to read, free to publish, typically under an open license like CC-BY.
Published in a fully open-access journal. The journal may charge article processing charges (APCs) to authors, but the paper is permanently free to read. Most gold OA papers carry an open license. You'll never lose access to a gold OA paper.
The journal itself is subscription-based (paywalled), but this specific article was made open access by the authors — usually by paying an APC. Hybrid papers always have an identifiable open license (e.g., CC-BY). The paper is permanently free even though most other articles in the same journal are not.
Currently free to read on the publisher's website, but no open license has been identified. This is the trickiest category. The publisher is letting you read it for free right now, but they haven't committed to keeping it that way.
Common reasons a paper ends up as bronze: the publisher made older issues free, it's a promotional free-read period, or the metadata simply doesn't include license information. Papertree shows a warning on bronze papers:
The paper is paywalled at the publisher, but a free copy exists in a repository — like arXiv, PubMed Central, or an institutional repository. The free copy might be the published version, an accepted manuscript, or a preprint (see "Versions" below). Green OA is common in fields like physics and computer science where preprint culture is strong.
No free version has been found anywhere. The paper is behind a paywall. Papertree will still show a link to the publisher's page where you can see the abstract, and you may be able to access it through your institution's subscription. The "View at publisher" button takes you to the DOI landing page.
A single paper can exist in multiple versions. Understanding the difference matters because the content can differ significantly between versions.
The final version of record — the one the journal published. It's been peer-reviewed, edited, formatted, and has the official page numbers and DOI. This is the definitive version and what you should cite.
Also called the "postprint" or "author accepted manuscript" (AAM). This version has passed peer review and includes all revisions, but hasn't been typeset or formatted by the publisher. The scientific content is identical to the published version — only the layout and page numbers differ. Many institutions require their researchers to deposit this version in a repository.
The version the authors submitted before or during peer review. This has not been peer-reviewed and may contain errors, incomplete analyses, or conclusions that changed in the final paper. Preprints are common on servers like arXiv and bioRxiv. They're useful for getting early access to research, but should be read with the understanding that the work hasn't been vetted yet.
Some papers in Papertree show a "Not yet published" label. This means the paper exists only as a preprint — it hasn't been accepted by or published in a peer-reviewed journal (at least not yet, or not that we know of). The research may still be under review, or the authors may have chosen not to submit it to a journal. Treat these with appropriate caution: the work has not been formally validated through peer review.
A retraction means the paper has been formally withdrawn by the authors or the journal, usually because of serious errors, data fabrication, plagiarism, or ethical issues. Retracted papers remain indexed (you can still find and read them), but their findings should not be relied upon. Papertree shows a prominent red warning for retracted papers, sourced from the Retraction Watch database via OpenAlex.
Some version cards in Papertree show a small license badge (like CC-BY or CC-BY-NC). These Creative Commons licenses tell you what you're allowed to do with the paper:
When no license badge appears, it means either the paper has no open license (bronze/closed) or the license information wasn't available in the metadata. Papers without a license are governed by default copyright — you can read them but redistribution and reuse are restricted.
All availability data comes from OpenAlex, an open catalog of the world's research that indexes over 250 million works. OpenAlex aggregates data from Crossref, DOAJ, PubMed, institutional repositories, and other sources to build a comprehensive picture of where each paper can be accessed.
Papertree processes this data as follows: