CERMINE review
A focused toolkit for extracting structured content and references from academic PDFs.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
CERMINE is an open-source Java library, standalone application, and web service for extracting structured information from academic PDF publications. It is aimed at researchers and teams that need article metadata, full text, references, or images in forms they can process further. Its outputs include NLM JATS XML, BibTeX, and plain text; it can also preserve reading order and label text zones by function. The Java/Maven API, REST service, and local batch processing make it relevant to both individual research workflows and larger document-processing pipelines.
Its strongest fit is scholarly content extraction rather than broad PDF automation. CERMINE parses metadata from reference and affiliation strings and can return parsed bibliographic references alongside article content. The web console is intended for born-digital PDFs, accepts files up to 25 MB, and is labeled experimental. OCR support is not included, so scanned documents that need text recognition are a poor fit. Teams should choose it when their PDFs are academic publications and structured scholarly output matters; general-purpose extraction or scanned-document workflows call for a different tool.
CERMINE is AGPL-licensed and open source, with cloud and self-hosted deployment options. The REST web service offers a way to access extraction remotely, while Java/Maven APIs and a standalone application support integration or local processing. The published project details describe a free, open-source tool rather than paid tiers, so the choice is primarily about technical fit and deployment. Documentation is the listed support channel. Researchers comfortable working with a Java-oriented tool should consider CERMINE for scholarly PDFs; organizations needing non-experimental service assurances or OCR should look elsewhere.
CERMINE pros and cons
- Where it wins
- Extracts metadata, full text, references, images, and reading-order text
- Outputs NLM JATS XML and supports BibTeX and plain text
- Offers Java/Maven and REST access, plus local batch processing
- Where it doesn't
- Designed for born-digital PDFs; OCR support is not included
- The web console is labeled experimental
- The focus is academic publications rather than general-purpose PDFs
CERMINE fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update