NBIB to BibTeX Converter
Free. No signup. Runs in your browser.
Upload or paste NBIB file and Convert to BibTeX. Continuation lines stay attached to their field, full author names (FAU) come through instead of just initials (AU), full journal titles (JT) come through instead of NLM abbreviations (TA), DOIs are extracted from LID / AID, and citekeys are generated as lastnameYearWord – ready to drop into your .bib without breaking the compile.
How to convert NBIB to BibTeX file online
thelatexlab.com/nbib-to-bibtex/ for next time..nbib file
From PubMed: select your results → Send to → Citation manager → Create file. You get a .nbib file (also called MEDLINE format). Drop it using the upload button, or paste the text directly. The parser handles single entries and large multi-entry exports separated by blank lines.Kr{\"a}mer) for older bibtex setups. Default is BibLaTeX..nbib content never leaves the tab. Entries that fail to parse show up inline with the reason so you can fix or drop them without losing the rest of the batch.Where most NBIB to BibTeX converters break
AU is “Last F” (initials only), FAU is “Last, First” (full names). Most converters take whichever comes first – usually AU, since it appears earlier in the record – and you end up with bibliographies that say “LeCun Y” instead of “LeCun, Yann”. The tool prefers FAU when present and falls back to AU only when it isn’t, matching what NLM intends.FAU over AU
full names preferredTA is the NLM-abbreviated title (N Engl J Med), JT is the full title (New England Journal of Medicine). Most bibstyles (APA, MLA, Chicago, IEEE) expect the full title and either render it full or abbreviate themselves using their own rules. The tool prefers JT when present so the source .bib stays in the expanded form. If your target style is Vancouver or a clinical journal that wants the NLM abbreviation, replace by hand or use a bibstyle option.JT over TA
full title preferredLID / AID
NBIB puts the DOI inside the LID or AID field along with a type marker: LID - 10.1038/nature14539 [doi] or as one entry among many in a multi-value AID list. Converters that don’t parse the marker either skip the DOI entirely or emit the whole [doi]-suffixed string as the doi field, which breaks resolution. The tool extracts the DOI proper, alongside the PMID and ISSN, so you get all three identifiers in the BibTeX. Acronyms in titles (DNA, HIV, COVID-19) are also brace-protected to survive case-folding bibstyles.436-44) expand to 436--444 when the result reads forward, matching what bibstyles expect. Trailing periods on titles are stripped (CSL/BibTeX convention – the citation style adds the period itself). Multi-value tags like AU, MH, KW, IS are kept as arrays rather than collapsed into a single comma-separated string.Frequently asked questions
NBIB (also called MEDLINE format) is the tagged plain-text export PubMed gives you via Send to → Citation manager. PubMed XML is a structured XML document with the same underlying record but a different shape, used mostly for programmatic ingestion (EFetch). This page reads NBIB. If you have PubMed XML and want BibTeX, the easiest path is to re-export from PubMed as NBIB, or – if you only have the PMID – paste it into /pubmed-to-bibtex/ for a direct lookup.
The tool prefers the full journal title (JT) when NBIB provides it. If the bibstyle then case-folds the title, the all-caps wrapping isn’t applied to journal fields by default (case-folding journal names is a bibstyle decision, not a converter one). If the source NBIB has only TA and not JT, the abbreviated form comes through. Brace-protect the abbreviation manually ({N. Engl. J. Med.}) if your style folds case and you want the abbreviation preserved.
Three common culprits: (1) the file is actually XML, not NBIB – check the first line; (2) the file is RIS or EndNote tagged format misnamed – /ris-to-bibtex/ and /endnote-to-bibtex/ handle those; (3) you exported from a UI that mangled the tag layout (turned 6-space continuations into 4-space, stripped the PMID line). Each failing entry shows up inline with the reason so you can fix the source.
Yes. The PMID stays alongside the DOI (extracted from LID or AID) and the ISSN (from the multi-value IS list, with the format marker stripped). Having all three means a downstream pipeline can re-resolve via whichever identifier is most stable for your context – the DOI for CrossRef, the PMID for biomedical reference managers.
Free, no signup. Parsing runs entirely in your browser – the .nbib file or pasted text never touches our server, no network call is made. The output exists in your browser session and disappears when you close the tab.
Yes – PubMed’s Send to → Citation manager is the standard export path for systematic-review pipelines (Covidence, Rayyan, RobotReviewer). The tool gives you a clean .bib ready for LaTeX. For thesis-scale workflows where the bibliography is part of a larger formatting job, see our thesis LaTeX formatting service, or email [email protected] with the project and target template.
You Have a Word or PDF File. We'll Convert it to Submission-Ready LaTeX.
We convert Word documents and PDFs into clean, submission-ready LaTeX for IEEE, Springer, Elsevier, ACM, arXiv, and more. Equations, tables, citations, and journal templates included. Compiled and tested in Overleaf before delivery.
Reviewed by a real LaTeX specialist.