API Overview
The Citation Wizard API provides three main services:
- Citation Generation - Convert URLs, DOIs, PDFs, and other content into properly formatted citations
- Research Parsing - Parse academic PDFs into structured scholarly JSON with GROBID and OpenAlex enrichment
- Paper Search - Search academic papers across CORE, Semantic Scholar, CrossRef, and UnPaywall databases
7f8a9b2c4d6e1f3a5b7c9d0e2f4a6b8c3e5f7a9b1c3d5e7f9a0b2c4d6e8f0a1b3c5Include in header:
X-API-Key or query parameter: api_key
Citation Generation API
Generate citations from URLs, DOIs, or other identifiers.
Request Body:
| Parameter | Type | Required | Description |
|---|---|---|---|
input |
string | Required | URL, DOI, PDF URL, or other identifier |
style |
string | Optional | Citation style (apa, ieee, mla, chicago, etc.). Default: apa |
format |
string | Optional | Output format: html, text, json, bibtex. Default: html |
Example Request:
The bibtex field is included in every response regardless of the requested format. You always get BibTeX alongside the formatted citation — no need to make a separate request.
Upload and process PDF, BibTeX, or RIS files for citation generation.
Form Data:
| Field | Type | Required | Description |
|---|---|---|---|
file |
file | Required | PDF, BibTeX (.bib), RIS (.ris), or text file |
style |
string | Optional | Citation style |
format |
string | Optional | Output format |
Example Request:
Try Citation API
Test the citation API directly. Enter a DOI, URL, arXiv ID, or upload a PDF.
Research Parsing API
New research-focused routes powered by GROBID for structured scholarly extraction. These routes are separate from the production citation APIs and are designed for API-first academic workflows.
Structured metadata, section hierarchy, extracted references, and optional OpenAlex enrichment using the polite pool email [email protected].
Upload a PDF and parse it into structured scholarly JSON using GROBID. Optionally enrich the parsed metadata with OpenAlex.
Form Data:
| Field | Type | Required | Description |
|---|---|---|---|
file |
file | Required | Academic PDF file to parse |
include_fulltext |
boolean | Optional | Include section hierarchy and references from fulltext TEI. Default: true |
include_raw_tei |
boolean | Optional | Include raw TEI XML in the response. Default: false |
enrich |
boolean | Optional | Enrich parsed metadata with OpenAlex. Default: true |
max_related |
integer | Optional | Number of related papers to return from the matched scholarly record. Default: 5, max: 10 |
Example Request:
Upload a PDF and extract bibliography references only.
Example Request:
Enrich a DOI or title using OpenAlex without uploading a file. Returns a resolved metadata record plus related papers.
Request Body:
| Parameter | Type | Required | Description |
|---|---|---|---|
doi |
string | Optional | DOI to enrich |
title |
string | Optional | Paper title to search when DOI is missing |
authors |
array | Optional | Optional author hints for better OpenAlex title matching |
max_related |
integer | Optional | Number of related papers to return. Default: 5, max: 10 |
Example Request:
Check research service readiness, including GROBID and OpenAlex.
Try Research API
Test the new research parsing endpoints directly from the docs.
NEW: AI-Powered Search Quality Tiers
The Paper Search API now supports AI-powered query enhancement using Azure OpenAI. Select a quality tier to improve search relevance.
| Tier | Model | Extra Latency | Description |
|---|---|---|---|
basic |
None | +0s | Rule-based preprocessing only (fastest) |
smart |
GPT-4o-mini | +1-3s | AI query reformulation, filter extraction |
intelligent |
GPT-4o | +5-10s | Deep understanding with domain expertise |
comprehensive |
GPT-4o | +15-30s | Multi-strategy search + AI reranking |
How AI Enhancement Works
- Query Reformulation: Natural language queries are converted to optimal academic search terms
- Filter Extraction: AI automatically detects year ranges, geographic focus, academic fields
- Domain Understanding: AI identifies the academic domain and adds relevant terminology
- Result Reranking: (comprehensive only) AI reorders results by relevance with explanations
Paper Search API
Search academic papers across multiple databases: CORE, Semantic Scholar, CrossRef, UnPaywall, and Google Scholar.
Search for academic papers across all providers with optional AI enhancement.
Request Body:
| Parameter | Type | Required | Description |
|---|---|---|---|
query |
string | Required | Search terms (natural language supported) |
limit |
integer | Optional | Results per page (1-100, default: 20) |
page |
integer | Optional | Page number (default: 1) |
year |
integer | Optional | Publication year filter |
sort |
string | Optional | Sort order: relevance, citationCount, year, year_old |
source |
string | Optional | Source: all, core, semantic, crossref, unpaywall, serper |
quality NEW |
string | Optional | AI quality tier: basic, smart, intelligent, comprehensive (default: basic) |
Example Request (Basic):
Example Request (AI-Enhanced - Smart):
Example Request (Comprehensive with Reranking):
Each paper includes a download_options array with pre-configured download methods. Simply POST the request_body to the endpoint to download the PDF.
Retrieve detailed information about a specific paper using its DOI.
Example Request:
Get information about available paper search sources and their capabilities.
Example Request:
Paper Search API (v2) - Advanced Filters
Enhanced search API with powerful filtering capabilities. Filter by open access, PDF availability, publishers, year ranges, and more.
- open_access - Filter by open access status (true/false)
- has_pdf - Only papers with downloadable PDF
- has_doi - Only papers with DOI
- publishers - Filter by Elsevier, Springer, Wiley, Nature, etc.
- year_from / year_to - Year range filter
- min_citations - Minimum citation count
- venues - Filter by journal name
- authors - Filter by author name
Search papers with advanced filtering options.
Request Body:
| Parameter | Type | Required | Description |
|---|---|---|---|
query |
string | Required | Search query |
limit |
integer | Optional | Results per page (1-100, default: 20) |
page |
integer | Optional | Page number (default: 1) |
source |
string | Optional | API source: all, core, semantic, crossref, unpaywall, serper (default: all) |
sort |
string | Optional | Sort: relevance, citationCount, year, year_old |
open_access |
boolean | v2 Filter | true = only open access, false = only non-OA, null = all |
has_pdf |
boolean | v2 Filter | true = only papers with PDF, false = only without |
has_doi |
boolean | v2 Filter | true = only papers with DOI |
publishers |
array | v2 Filter | Filter by publishers: ["elsevier", "springer", "wiley", "nature"] |
year_from |
integer | v2 Filter | Papers from this year onwards |
year_to |
integer | v2 Filter | Papers up to this year |
min_citations |
integer | v2 Filter | Minimum citation count |
venues |
array | v2 Filter | Filter by journal names: ["nature", "lancet"] |
authors |
array | v2 Filter | Filter by author names |
Example Request - Open Access with PDF:
Example Request - Specific Publishers:
Search papers using GET parameters (simpler filters only).
Get documentation of all available filter options, including common publisher names.
elsevier, springer, wiley, nature, oxford,
cambridge, taylor, sage, ieee, acm,
plos, bmc, mdpi, frontiers, hindawi
PDF Download API (v2)
Download academic papers using multiple methods with automatic fallback. The API tries each method in order until one succeeds.
- Unpaywall - Open access papers (requires DOI)
- Fulltext - Elsevier, Springer, Wiley, PLOS, etc. (requires DOI)
- PubMed - PubMed Central articles (requires PMCID)
- Direct - Direct PDF URLs from search results
- SciDownl - Last resort fallback (requires DOI)
Smart download that automatically tries multiple methods with fallback until one succeeds.
Request Body:
| Parameter | Type | Required | Description |
|---|---|---|---|
doi |
string | Optional* | DOI for unpaywall, fulltext, scidownl methods |
pmcid |
string | Optional* | PubMed Central ID (e.g., PMC1234567) for pubmed method |
pdf_url |
string | Optional* | Direct PDF URL for direct method |
filename |
string | Optional | Custom filename (without extension) |
methods |
array | Optional | Ordered list of methods to try: ["unpaywall", "fulltext", "pubmed", "direct", "scidownl"] |
stop_on_success |
boolean | Optional | Stop after first success (default: true) |
* At least one identifier (doi, pmcid, or pdf_url) is required
Example Request:
Individual Download Methods
You can also call each download method directly:
Download via Unpaywall (open access papers). Requires: doi
Download via Fulltext Downloader (Elsevier, Springer, Wiley, PLOS, etc.). Requires: doi
Download from PubMed Central. Requires: pmcid
Download from direct PDF URL. Requires: pdf_url
Download via SciDownl/Sci-Hub (last resort). Requires: doi
Get information about all available download methods, their requirements, and capabilities.
Download a previously downloaded PDF file. The filename is returned in the download response.
Paper Search Data Sources
Our paper search API aggregates results from five major academic databases:
CORE API
Coverage: 30+ million open access papers from repositories worldwide
Strengths: Open access focus, repository metadata, full-text search
Rate Limit: 10 requests/second
Semantic Scholar
Coverage: Computer science, biomedical, and general academic papers
Strengths: Citation analysis, author information, paper relationships
Rate Limit: 100 requests/5 minutes
CrossRef API
Coverage: Academic publications with DOIs
Strengths: Authoritative metadata, publisher information
Rate Limit: 50 requests/second
UnPaywall API
Coverage: Open access versions of paywalled papers
Strengths: Open access detection, PDF links
Rate Limit: 100,000 requests/day
Google Scholar (via Serper)
Coverage: Broad academic coverage including theses, books, conference papers
Strengths: Comprehensive search, PubMed/PMC links, citation data
Rate Limit: API key required
Source ID: serper
Try the API
Test Paper Search
Try searching for academic papers using our unified API with AI enhancement:
Utility Endpoints
Check system health and service availability.
Get available citation styles and their metadata.