Skip to main content
For Developers
EXPLORE for Developers

The graph is open.
Build on it.

Every record in the OpenAIRE Graph, 400M+ research products linked to researchers, organisations, funders, projects, datasets and software, is available to build with. Query it live through the API, download it in bulk, run it at scale through a data warehouse, or connect an AI agent directly to it. This page covers every route in.

OpenAIRE Graph APIs

The OpenAIRE Graph API lets you query research products, organisations, data sources, projects and persons directly, with search, filtering and sorting built in. No key is required to start, authenticate only when you need a higher rate limit.

  • Graph API
    Query research products, organizations, data sources, projects and persons directly over REST.
  • RDA SKG-IF API
    Implements the RDA Scholarly Knowledge Graph Interoperability Framework, a shared specification also adopted by other knowledge graphs. Build against it once and work across more than just OpenAIRE.
  • ScholeXplorer API
    Returns dataset-to-publication and dataset-to-dataset links, the citation-style relationships that connect research data to the literature and to other datasets.
  • DSpace & EPrints API
    Tailored, funder-scoped metadata access built specifically for DSpace and EPrints repository platforms, covering a selection of international funders.
  • Broker API
    The interface repositories, publishers and aggregators use to receive enrichment suggestions mined from full text, citations, funding mentions, related-resource links, and write accepted ones back into their own metadata.
Full documentation
Swagger UI

Bulk download on Zenodo

Need the whole graph rather than live queries? Bi-annual snapshots of the full OpenAIRE Graph are published on Zenodo as JSON and Parquet, openly licensed under CC-BY, no negotiations, no gatekeeping.

  • Bi-annual snapshots, citable DOI
    Each snapshot is assigned its own DOI, so your work built on a specific version of the graph remains reproducible and citable.
  • CC-BY, no restrictions
    Use it for research, build a commercial product on top of it, redistribute it, the only requirement is attribution.
  • JSON and Parquet formats
    Parquet is well suited to large-scale analytics workloads, JSON is easier to parse for smaller or exploratory projects.
Graph snapshots on Zenodo

Cloud access

The OpenAIRE Graph is published as a public dataset on Google Cloud, queryable directly from BigQuery. No download, no ingestion pipeline of your own to build or maintain.
  • No ETL required
    Query with standard SQL directly in BigQuery Studio, one table per entity type, with a Google account that's all you need to get started.
  • Storage is on us, processing is on you
    OpenAIRE covers the data storage costs. Google grants $300 in credit for first-time users, and a query shuffling 1TB typically costs around $6, plenty for experimentation.

Separately, ORION DBs is an independent, community-run collective hosting open research datasets on BigQuery, several versions of the OpenAIRE Graph included, alongside Crossref, OpenAlex, DataCite, ORCID and more. Because everything lives in the same environment, you can join OpenAIRE data with these other sources directly in a single query, and get started with no billing account via the free BigQuery sandbox.

Getting started on Google cloud
Explore ORION DBs

Access via AI agents

OpenAIRE has partnered with Alien Intelligence, a french startup building agentic AI systems, to connect this agent layer directly to the OpenAIRE Graph.

Grounded, not guessing. Instead of relying on general web content or text prediction alone, agents built on this connection work with authoritative scholarly metadata, persistent identifiers and verified relationships. Every answer can be traced back to its source in a graph that is rebuilt regularly, with deduplication and validation of ORCID and ROR identifiers built in.

What becomes possible

  • Literature reviews across a field
  • Citation analysis over time
  • Author and career mapping through ORCID
  • Dataset discovery across repositories
  • Bibliometric signals on emerging topics
  • Cross-domain links between adjacent fields
  • ... and much more

Interested in joining the testing group? Contact This email address is being protected from spambots. You need JavaScript enabled to view it. to get involved.