DBLP Has a Free API — Search Every Computer Science Paper Ever Published

#api #python #computerscience #tutorial

Looking for a specific CS paper? Or all publications by a researcher? Or papers from a specific conference?

DBLP indexes virtually every computer science publication — 6.5M+ papers from conferences, journals, and workshops. And it has a free API. No key needed.

What Is DBLP?

DBLP (Digital Bibliography & Library Project) is THE reference database for computer science. It indexes:

6.5M+ publications (papers, articles, books)
3.5M+ authors with disambiguated profiles
All major CS conferences (NeurIPS, ICML, CVPR, ACL, SIGMOD, etc.)
All major CS journals (JACM, IEEE TPAMI, VLDB Journal, etc.)

Quick Start

import requests

# Search for papers
resp = requests.get("https://dblp.org/search/publ/api", params={
    "q": "attention is all you need",
    "format": "json",
    "h": 5
})

hits = resp.json()["result"]["hits"]["hit"]
for hit in hits:
    info = hit["info"]
    title = info.get("title", "N/A")
    year = info.get("year", "N/A")
    venue = info.get("venue", "N/A")
    authors = info.get("authors", {}).get("author", [])
    if isinstance(authors, dict): authors = [authors]
    author_names = [a.get("text", a) if isinstance(a, dict) else a for a in authors[:3]]
    print(f"[{year}] {title}")
    print(f"  Venue: {venue} | Authors: {', '.join(author_names)}")
    print()

Find All Papers by an Author

# Search for author
resp = requests.get("https://dblp.org/search/author/api", params={
    "q": "Yann LeCun",
    "format": "json",
    "h": 3
})

hits = resp.json()["result"]["hits"]["hit"]
for hit in hits:
    name = hit["info"]["author"]
    url = hit["info"]["url"]
    print(f"{name}: {url}")

# Get all papers by author URL key
author_key = "l/YannLeCun"  # From DBLP author URL
resp = requests.get(f"https://dblp.org/pid/{author_key}.xml", params={"format": "json"})

Search by Venue (Conference/Journal)

# Find papers from NeurIPS 2024
resp = requests.get("https://dblp.org/search/publ/api", params={
    "q": "venue:NeurIPS year:2024 transformer",
    "format": "json",
    "h": 5
})

hits = resp.json()["result"]["hits"]["hit"]
for hit in hits:
    info = hit["info"]
    print(f"[{info.get('year')}] {info.get('title')}")

DBLP vs Other CS Paper APIs

Feature	DBLP	Semantic Scholar	OpenAlex	Google Scholar
Focus	CS only	All fields	All fields	All fields
Papers	6.5M+ CS	200M+ all	250M+ all	Unknown
API Key	No	Yes (free)	No	No API
Author profiles	Disambiguated	Yes	Yes	No
Conference data	Excellent	Good	Good	No
Completeness for CS	Best	Good	Good	Incomplete