Reference Converter
Convert identifiers and bibliography files to APA, MLA, BibTeX, RIS and more. First 10 references free; review fields before copying or downloading.
Your references BibTeX
Files are processed in your browser. Identifier lookups use metadata services. Drafts and results are saved in this tab for this page; the allowance count is saved locally.
Continue your research in AnswerThis
Find papers and develop your research with linked sources. Opens in a new tab; download your bibliography first. Files and references do not transfer automatically.
Choose between identifier lookup and file conversion
The reference converter supports two starting points. Identifiers such as DOI, PMID, arXiv ID and ISBN need metadata lookup before they can become references. Bibliography files already contain named or tagged fields and can be parsed in your browser. Select the format that describes your actual input, then choose the destination. A file extension alone cannot turn a formatted reference list into BibTeX or make a PMCID into a PMID.
Paste one identifier or plain reference per line, or upload a supported bibliography file. Review returned records before copying or downloading. Identifier batches can partially succeed, so keep failed lines for follow-up. File conversion depends on the quality of the exported metadata and cannot recover fields absent from the file. Keep the original export until the destination library or document has been checked.
Which output format fits your next step?
APA 7 and MLA 9 produce readable references for a document. BibTeX is a structured database suited to compatible LaTeX workflows. RIS is tagged interchange for reference managers. EndNote XML and NBIB support destination workflows that explicitly accept those formats, while CSV makes fields easier to inspect in a spreadsheet. DOI output extracts identifiers and requires a recorded DOI; it is not a complete bibliography.
Choose the destination by what it actually imports. A plain-text reference may look correct but does not retain a manager’s editable field structure. Conversely, a .bib file can contain excellent metadata without looking like the final reference list. The manuscript’s style processor determines the final display. Imported attachments, folder membership and existing manuscript citation links are separate from bibliographic interchange.
Input → output
Input
@book{river2020, author={River, Alex}, title={Research notes}, year={2020}}
Output
TY - BOOK AU - River, Alex TI - Research notes PY - 2020 ER -
Synthetic book fixture; minimal mapping, not exact export serialization.
Worked example: a small library transfer
The synthetic example below converts one BibTeX book into a minimal RIS record. A book entry becomes TY BOOK, its title becomes TI, its person becomes AU and its year becomes PY. ER closes the record. This illustrates how a converter moves fields, rather than merely changing punctuation. The example’s author and title are invented for testing and should not be cited as a real publication.
Check one book, one article and any unusual source type from your own file before importing the entire result. Compare creator counts, title, year, identifiers and container details. For a plain reference list, parsing depends on recognizable patterns; unread fields can remain empty. The original line may be retained in a BibTeX note for comparison, but that does not guarantee every bibliographic field was interpreted correctly.
Provider coverage, losses and a practical review pass
DOI lookups use Crossref and, when Crossref has no record, DataCite. PubMed identifiers use NCBI PubMed metadata, ISBNs use Open Library and arXiv identifiers use arXiv. BibTeX, RIS, EndNote XML, NBIB and CSV are parsed locally; no AI writes missing bibliographic facts. Plain-text parsing is also a pattern-based operation, so unusual reference layouts need close inspection rather than confidence based on a polished-looking output.
Formats differ in supported fields, name conventions and source types. Compare corporate authors, accents, abstracts, keywords, identifiers and custom fields after conversion, and read any warnings about losses. Keep library attachments separately. Import into a test collection first, resolve duplicates and only then combine it with an established bibliography. A downloaded file is a starting point for review, not evidence that every destination field survived unchanged.
Conversion limits and processing
- The first 10 references are free per browser, shared across all six converter pages. Files: up to 1 MB; batches: up to 100 references.
- File formats are parsed in your browser. Identifier lookups go through the server to Crossref / DataCite (DOI), NCBI PubMed (PMID), Open Library (ISBN) or arXiv. No AI is used to convert references.
- Review any field-loss warnings and keep your original file. Attachments and custom library fields are not a complete portable backup.
Popular reference conversions
Frequently asked questions
Which input formats are supported?
Choose DOI, PubMed ID, arXiv ID, ISBN, BibTeX, RIS, EndNote XML, NBIB, CSV or a plain reference list. Keep one identifier or prose reference per line. Files and identifiers use different processing paths, so select the actual input format.
How do I choose the output format?
Use APA or MLA for readable references, BibTeX for a compatible LaTeX database and RIS for reference-manager interchange. Select EndNote XML, NBIB or CSV when your destination expects it. DOI output needs a recorded DOI and contains identifiers rather than complete references.
How much conversion is free?
The first 10 references are free without login, shared across all six converter pages in this browser. File uploads are limited to 1 MB and batches to 100 references. Copying and downloading an existing result do not consume additional references.
Can the converter fill every missing field?
No. Identifier providers may supply incomplete metadata, and file parsing can only use what the file contains. Review original publications for missing dates, creators or identifiers. Conversion does not use AI to invent bibliography details or determine whether a source supports a claim.
How should I recover from a malformed file?
Check entry boundaries and field syntax, or export the bibliography again from its source application. For CSV, use a header row and quote values containing commas. Test a small complete record before retrying a large batch with the same structural problem.
Does conversion preserve attachments and custom fields?
No complete library transfer is promised. Attachments, folders, custom fields, citation keys and formatting may not map between formats. Read loss warnings, retain the original file and inspect imported records before replacing an existing library or manuscript bibliography.
What happens after I download the result?
Import it using your destination application’s supported file workflow or copy readable references into a document. Check ordering and layout in that destination. Opening AnswerThis starts a separate research session; the converter does not upload or transfer your bibliography automatically.