Citation Formatter: DOI to APA, MLA, Chicago, Harvard, IEEE, Vancouver, BibTeX & RIS

Paste DOIs, ISBNs, arXiv IDs, titles or a messy reference list. Get clean citations in six styles plus BibTeX and RIS, with a flag on anything that needs a second look.

Formatting references by hand is slow, and online generators guess. Citation Formatter looks every reference up in the official registries (Crossref, DataCite, Open Library, plus PubMed for PMIDs), then formats it with citeproc-js, the open-source citation engine behind Zotero and Mendeley, using the official CSL style files. Nothing is hand-rolled, so the commas, italics and initials follow the style guide.

• 📚 Six styles in one run: APA 7, MLA 9, Chicago author-date (18th ed.), Harvard (Cite Them Right 12th ed.), IEEE and Vancouver (NLM)
• 🔁 Exports for reference managers: BibTeX (.bib) for LaTeX and Overleaf, and RIS (.ris) for Zotero, Mendeley, EndNote and Rayyan
• 🧭 Takes whatever you have: DOIs, DOI links, ISBNs, arXiv IDs, PMIDs, bare titles, references in any style, or a pasted .bib/.ris file
• 🚦 Honest confidence: every reference is marked OK, CHECK or FIX, with a plain-language reason
• 🧹 Cleans up registry data: sentence-case titles for APA, initials fixed ("AJ" → "A. J."), series vs. book title for conference chapters, duplicates removed
• 💸 $0.003 per formatted citation. All six styles, BibTeX and RIS are included in that price. References we can't find are free.

Who it's for

• Researchers, grad students and postdocs: convert a bibliography from one journal's style to another before submission.
• Editors, proofreaders and thesis formatters: turn a messy reference list into a clean APA or Harvard list in seconds, then spend your time on the edge cases we flag.
• Librarians and research support teams: batch-convert reading lists and export them to Zotero or EndNote.
• Developers and AI agents: turn DOIs into formatted citations through the Apify API or MCP, without maintaining a citation engine.

How to use it

1. Click Try for free. Put one reference per line in References, or paste a whole list (or a .bib/.ris file) into Paste a reference list.
2. Pick your styles (all six are on by default and cost the same).
3. Click Start. Ten references usually take 10–20 seconds (preprint titles without a DOI can take up to a minute while we search DataCite).
4. Download the ready-to-paste bibliography from Storage → Key-value store (BIBLIOGRAPHY_APA, BIBLIOGRAPHY_APA_HTML with italics for Word or Google Docs, BIBLIOGRAPHY_BIBTEX, BIBLIOGRAPHY_RIS...), or open the Formatted citations table.
5. Open Needs review to see the CHECK and FIX items and why they were flagged.

Input example

{
  "references": [
    "10.1038/nature14539",
    "ISBN 978-0-262-03384-8",
    "arXiv:1706.03762",
    "PMID: 26017442",
    "He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. CVPR 2016",
    "https://doi.org/10.1016/S0140-6736(97)11096-0"
  ],
  "text": "@article{watson1953,\n  author = {Watson, J. D. and Crick, F. H. C.},\n  title = {Molecular Structure of Nucleic Acids},\n  journal = {Nature}, volume = {171}, number = {4356},\n  pages = {737--738}, year = {1953}, doi = {10.1038/171737a0}\n}",
  "styles": ["apa", "mla", "chicago", "harvard", "ieee", "vancouver"],
  "includeBibtex": true,
  "includeRis": true
}

Output example

Real output from a platform run:

APA 7

LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539

Cormen, T. H., Leiserson, C. E., Rivest, R. L., & Stein, C. (2009). Introduction to algorithms (3rd ed.). The MIT Press.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention is all you need (arXiv:1706.03762). arXiv. https://doi.org/10.48550/ARXIV.1706.03762

MLA 9

LeCun, Yann, et al. “Deep Learning.” Nature, vol. 521, no. 7553, May 2015, pp. 436–44, https://doi.org/10.1038/nature14539.

Chicago author-date

LeCun, Yann, Yoshua Bengio, and Geoffrey Hinton. 2015. “Deep Learning.” Nature 521 (7553): 436–44. https://doi.org/10.1038/nature14539.

Harvard (Cite Them Right)

LeCun, Y., Bengio, Y. and Hinton, G. (2015) ‘Deep learning’, Nature, 521(7553), pp. 436–444. Available at: https://doi.org/10.1038/nature14539.

IEEE

[1] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2016, pp. 770–778. doi: 10.1109/cvpr.2016.90.

Vancouver (NLM)

1. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015 May 27;521(7553):436–44. doi:10.1038/nature14539

One dataset row per reference (shortened):

{
  "index": 1,
  "input": "10.1038/nature14539",
  "inputType": "doi",
  "status": "OK",
  "confidence": 1,
  "matchMethod": "doi",
  "doi": "10.1038/nature14539",
  "title": "Deep learning",
  "authors": "LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey",
  "year": 2015,
  "apa": "LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539",
  "mla": "LeCun, Yann, et al. “Deep Learning.” Nature, vol. 521, no. 7553, May 2015, pp. 436–44, https://doi.org/10.1038/nature14539.",
  "inText": {"apa": "(LeCun et al., 2015)", "mla": "(LeCun et al.)", "harvard": "(LeCun, Bengio and Hinton, 2015)", "ieee": "[1]"},
  "bibtex": "@article{LeCun2015Deep, ...}",
  "ris": "TY  - JOUR ...",
  "warnings": []
}

What OK, CHECK and FIX mean

Status	Meaning	Charged?
🟢 OK	Exact DOI/ISBN lookup, or a text match with 90%+ confidence	Yes
🟡 CHECK	Formatted, but read the note first: the match was by title only, the venue or version differs (e.g. arXiv vs. proceedings), the record is retracted, or the registry is missing authors or a date	Yes (duplicates are free)
🔴 FIX	Not found anywhere. Often a typo, a non-DOI source, or a fabricated reference. Listed at the end of each bibliography file as [UNRESOLVED - check manually]	No

Retracted papers are flagged in CHECK. To screen a whole bibliography for fabricated citations, retractions and wrong details, run our sister Actor, Reference Checker: Fake Citation & Retraction Detector.

Files you get

Key-value store key	What it is
BIBLIOGRAPHY_APA, _MLA, _CHICAGO, _HARVARD, _IEEE, _VANCOUVER	Plain-text bibliography per style, sorted the way the style requires (alphabetical or numbered)
BIBLIOGRAPHY_<STYLE>_HTML	The same list with italics and hanging indents. Open it and paste into Word or Google Docs
BIBLIOGRAPHY_BIBTEX	Combined .bib file
BIBLIOGRAPHY_RIS	Combined .ris file
BIBLIOGRAPHY_CSL_JSON	Cleaned CSL-JSON, for Pandoc, Zotero or your own code
SUMMARY	Counts by status and the list of files

Pricing

$0.003 per formatted citation (pay-per-event). Every selected style, BibTeX, RIS and in-text citations are included.

References	Cost
50 (a typical paper)	$0.15
300 (a thesis)	$0.90
1,000	$3.00

Unresolved references and duplicates are never charged. You can also set a maximum charge per run in Apify.

FAQ

How accurate is the formatting? Formatting is done by citeproc-js with the official CSL styles, the same engine and styles used by Zotero and Mendeley. We checked the output against hand-verified examples from the APA, MLA, Chicago, Cite Them Right, IEEE and NLM guides. The weak spot of every citation tool is the source data, so we fix the common registry problems: Title Case titles, bare initials, and the series printed in place of the book title.

Why did my title change case? APA, Harvard, IEEE and Vancouver use sentence case. Registries usually store Title Case, so we convert it, keeping acronyms (CVPR, mRNA), mixed-case words (iPhone) and names like Covid-19 or U-Net. A tool can't always tell a proper noun from an ordinary word, so check converted titles for names and places (each one carries a note). You can turn this off with Sentence-case titles.

What about books, preprints and datasets? Books resolve by ISBN through Open Library, including the edition. arXiv preprints resolve through their DataCite DOIs, and datasets and software on Zenodo, Dryad or Figshare resolve by DOI.

Can it handle references without a DOI? Yes. Free-text references and titles are matched against Crossref, OpenAlex and DataCite. A match needs a high title, author and year score, and weaker matches are refused rather than guessed (FIX). Websites, newspaper articles and grey literature without a DOI or ISBN usually end up as FIX.

Can I call it from code or an AI agent? Yes, through the Apify API, the Python or JavaScript clients, or Apify's MCP server. Send DOIs and get formatted citations back in one call.

Does it scrape websites? No. It only uses official public APIs (Crossref, DataCite, Open Library, NCBI E-utilities, OpenAlex), follows their rate limits and identifies itself politely.

Data sources and credits

• Crossref and DataCite metadata via DOI content negotiation, plus the Crossref REST API
• Open Library for ISBNs, NCBI E-utilities for PMIDs, and OpenAlex for title matching
• citeproc-js and citation.js for formatting and BibTeX/RIS
• Styles and locales from the CSL project (CC BY-SA 3.0)

---

Built by Panda Federation

Self-healing, local-first AI systems and data tools that run themselves.

Panda Federation builds small, dependable tools that do one job well and keep doing it. Citation Formatter is the second module in our research toolkit, next to Reference Checker.

More from Panda Federation on Apify · GitHub