bibr

bibr reads scientific papers (PDF) and extracts their structure: title, authors and affiliations, abstract, sections, references, figures and tables, as machine-readable JSON.

This is the hosted API front door. Sign in to upload a paper in the browser or to create a personal API token for scripts and for Pytacheck.

Sign in

Privacy

Files you upload are processed on project servers in the EU/EEA. Page images are sent to a rented GPU (RunPod, Iceland) for text recognition. Uploads are deleted once they have been read, results are kept for 1 hour and then discarded, and nothing you upload or extract is used for training.

Access is limited to invited testers. Sign-in is handled by auth.metacheck.app.