Licensed legal data for AI training, RAG and evals European law, structured for machines.
EU legislation, CJEU case law and national legislation from 15 countries, parsed to article level, versioned by content hash and linked by a typed citation graph. Every record carries its source, licence and attribution. From Arieta Labs, the company behind LexAPI.
We reply within one business day. Custom cuts per jurisdiction, language or date range in 5–6 business days.
- legal documents in the corpus
- 3.9M+
- citations in the corpus graph
- 5.4M+
Ready to license today
- documents
- 410k+
- articles
- 4M+
- tokens
- 2B+
- citation edges
- 800k+
24 languages · 16 jurisdictions · 1801–2026 · counts as of the November 2026 snapshot, rounded down.
Plus 3.4M+ national court decisions, delivered with personal data masked. Tell us which courts you need
5.4M+ citations in the corpus · every edge typed
What’s in the box
The text is free. The structure, the history and the licensing work are not.
Three datasets cut from one snapshot, plus eval sets built to your spec. They share ids, jurisdiction codes, content hashes and the snapshot id. Parquet (zstd) and gzip JSON Lines.
Corpus
Article-level EU and national legal text with full metadata.
Instead of: writing article-level parsers for 20+ publication formats.
410k+documents
History
Every content-hash version we have observed, for point-in-time training and evals.
Instead of: a time machine. Official sites show versions, not what they published on a given day.
15k+versions
Citation graph
Typed citation edges across EU and national law.
Instead of: resolving citations, amendments and transpositions by hand.
800k+edges
Eval
Retrieval QA and citation QA sets, built to your spec from the graph and the history.
Instead of: building gold sets from scratch. Tell us your tasks and jurisdictions.
Built to your spec
Why version history matters for evals → Citation graphs as supervision →
Need a custom cut? Per jurisdiction, language, date range or document type, delivered in 5–6 business days.
Record explorer
Look at the rows before you talk to us.
Sample records
{
"id": "32016R0679:en",
"celex": "32016R0679",
"jurisdiction": "EU",
"language": "en",
"kind": "legislation",
"title": "Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (Text with EEA relevance)",
"documentType": "Regulation",
"author": "Council of the European Union, European Parliament",
"dateOfDocument": "2016-04-27",
"dateOfEffect": "2016-05-24",
"dateOfPublication": null,
"ecli": null,
"eli": "http://data.europa.eu/eli/reg/2016/679/oj",
"subjects": [
"Area of freedom, security and justice",
"Consumer protection",
"Information and verification"
],
"keywords": [
"access to information",
"area of freedom, security and justice",
"cross-frontier data flow",
"data protection",
"data-processing law",
"disclosure of information",
"natural person",
"personal data",
"protection of privacy"
],
"version": 14,
"contentHash": "3c91009d05bbda7b9717678970c2d4d680cee402b00f8f0df92510f9bab42543",
"recitals": [
"The protection of natural persons in relation to the processing of personal data is a fundamental right. Article 8(1) of the Charter of Fundamental Rights of the European Union (the ‘Charter’) and Article 16(1) of the Treaty on the Functioning of the European Union (TFEU) provide that everyone has the right to the protection of personal data concerning him or her.",
"The principles of, and rules on the protection of natural persons with regard to the processing of their personal data should, whatever their nationality or residence, respect their fundamental rights and freedoms, in particular their right to the protection of personal data. This Regulation is intended to contribute to the accomplishment of an area of freedom, security and justice and of an economic union, to economic and social progress, to the strengthening and the convergence of the economies within the internal market, and to the well-being of natural persons."
],
"articles": [
{
"id": "art_1",
"number": "Article 1",
"title": "Subject-matter and objectives",
"text": "1. This Regulation lays down rules relating to the protection of natural persons with regard to the processing of personal data and rules relating to the free movement of personal data. 2. This Regulation protects fundamental rights and freedoms of natural persons and in particular their right to the protection of personal data. 3. The free movement of personal data within the Union shall be neither restricted nor prohibited for reasons connected with the protection of natural persons with regard to the processing of personal data.",
"repealed": false
},
{
"id": "art_2",
"number": "Article 2",
"title": "Material scope",
"text": "1. This Regulation applies to the processing of personal data wholly or partly by automated means and to the processing other than by automated means of personal data which form part of a filing system or are intended to form part of a filing system. 2. This Regulation does not apply to the processing of personal data: (a) in the course of an activity which falls outside the scope of Union law; (b) by the Member States when carrying out activities which fall within the scope of Chapter 2 of Title V of the TEU; (c) by a natural person in the course of a purely personal or household activity; (d […]",
"repealed": false
},
{
"id": "art_3",
"number": "Article 3",
"title": "Territorial scope",
"text": "1. This Regulation applies to the processing of personal data in the context of the activities of an establishment of a controller or a processor in the Union, regardless of whether the processing takes place in the Union or not. 2. This Regulation applies to the processing of personal data of data subjects who are in the Union by a controller or processor not established in the Union, where the processing activities are related to: (a) the offering of goods or services, irrespective of whether a payment of the data subject is required, to such data subjects in the Union; or (b) the monitoring […]",
"repealed": false
},
{
"id": "art_4",
"number": "Article 4",
"title": "Definitions",
"text": "For the purposes of this Regulation: (1) ‘personal data’ means any information relating to an identified or identifiable natural person (‘data subject’); an identifiable natural person is one who can be identified, directly or indirectly, in particular by reference to an identifier such as a name, an identification number, location data, an online identifier or to one or more factors specific to the physical, physiological, genetic, mental, economic, cultural or social identity of that natural person; (2) ‘processing’ means any operation or set of operations which is performed on personal data […]",
"repealed": false
},
{
"id": "art_5",
"number": "Article 5",
"title": "Principles relating to processing of personal data",
"text": "1. Personal data shall be: (a) processed lawfully, fairly and in a transparent manner in relation to the data subject (‘lawfulness, fairness and transparency’); (b) collected for specified, explicit and legitimate purposes and not further processed in a manner that is incompatible with those purposes; further processing for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes shall, in accordance with Article 89(1), not be considered to be incompatible with the initial purposes (‘purpose limitation’); (c) adequate, relevant and limited t […]",
"repealed": false
}
],
"source": {
"name": "EUR-Lex / Cellar (Publications Office of the European Union): EU law and the Official Journal",
"url": "https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32016R0679",
"licence": "EUR-Lex legal notice (reuse authorised, source acknowledged; Commission documents under Decision 2011/833/EU)",
"licenceUrl": "https://eur-lex.europa.eu/content/legal-notice/legal-notice.html",
"attribution": "Source: EUR-Lex, © European Union, 1998–2026; reuse authorised under the EUR-Lex legal notice (Commission documents: Decision 2011/833/EU). Changes made by Arieta Labs: text extracted from the published documents, restructured into articles and metadata, citation edges derived. Only the Official Journal is authentic."
},
"snapshotId": "2026-11"
}Coverage
EU law plus 15 countries.
EU legislation, CJEU case law and national legislation ship in every snapshot. License all of it, or only the lanes, languages and years you need.
- tokens
- 2B+
- date range
- 1801–2026
- languages
- 24
- lanes with records
- 16
3.4M+ national court decisions: included, delivered with personal data masked. Email us
In preparation: Portugal, Belgium, Slovakia, Hungary, Bulgaria, Romania, Greece, Latvia, Estonia. See coverage
All lanes, licences and exclusions →Want one country, or one language? Tell us the lanes.
Show as table
| Year | Documents |
|---|---|
| before 1951 | 2,645 |
| 1951 | 97 |
| 1952 | 88 |
| 1953 | 85 |
| 1954 | 86 |
| 1955 | 110 |
| 1956 | 93 |
| 1957 | 103 |
| 1958 | 165 |
| 1959 | 131 |
| 1960 | 118 |
| 1961 | 138 |
| 1962 | 164 |
| 1963 | 114 |
| 1964 | 229 |
| 1965 | 225 |
| 1966 | 185 |
| 1967 | 249 |
| 1968 | 276 |
| 1969 | 285 |
| 1970 | 373 |
| 1971 | 335 |
| 1972 | 438 |
| 1973 | 431 |
| 1974 | 425 |
| 1975 | 497 |
| 1976 | 453 |
| 1977 | 589 |
| 1978 | 560 |
| 1979 | 596 |
| 1980 | 619 |
| 1981 | 637 |
| 1982 | 917 |
| 1983 | 852 |
| 1984 | 874 |
| 1985 | 906 |
| 1986 | 1,012 |
| 1987 | 2,642 |
| 1988 | 2,797 |
| 1989 | 2,798 |
| 1990 | 3,959 |
| 1991 | 4,630 |
| 1992 | 5,446 |
| 1993 | 5,150 |
| 1994 | 5,258 |
| 1995 | 5,298 |
| 1996 | 5,978 |
| 1997 | 5,569 |
| 1998 | 5,381 |
| 1999 | 5,601 |
| 2000 | 5,577 |
| 2001 | 6,537 |
| 2002 | 6,706 |
| 2003 | 12,426 |
| 2004 | 14,003 |
| 2005 | 26,762 |
| 2006 | 26,489 |
| 2007 | 26,612 |
| 2008 | 28,146 |
| 2009 | 33,427 |
| 2010 | 92,864 |
| 2011 | 101,795 |
| 2012 | 108,678 |
| 2013 | 137,121 |
| 2014 | 148,599 |
| 2015 | 150,657 |
| 2016 | 163,792 |
| 2017 | 155,755 |
| 2018 | 141,391 |
| 2019 | 128,702 |
| 2020 | 143,641 |
| 2021 | 311,733 |
| 2022 | 432,948 |
| 2023 | 386,839 |
| 2024 | 353,809 |
| 2025 | 385,471 |
| 2026 | 268,488 |
By the date of each document. 40k+ documents without a date are not shown.
Build or buy
The questions every data team asks first.
Short answers, with the evidence one click away.
-
Why not scrape EUR-Lex ourselves?
The text is free; the work is not. This is 16 jurisdictions and 20+ publication formats parsed to article level, 800k+ citation edges resolved and typed, withdrawals tracked as tombstones, and the reuse terms documented source by source. You get it as one schema, maintained monthly.
Build vs buy, in detail -
Is it licensed for AI training?
Every record carries its upstream source, licence and attribution. Rights for RAG, fine-tuning, pre-training and model weights are set by tier in one licence agreement. What cannot be licensed for AI, such as UK case law, is left out and explained.
Rights matrix -
How fresh is it?
Monthly delta snapshots: new and changed records as upserts, withdrawn ones as tombstones. The current snapshot runs to 12 November 2026. Version history records when each change was observed.
Version history -
What formats?
Apache Parquet (zstd, about 512 MB shards) and gzip JSON Lines with identical fields. Each snapshot ships MANIFEST.json with a SHA-256 per shard, a datasheet and the attribution file; Croissant metadata is public.
Formats and files -
How is it delivered?
By signed download or straight to your bucket. Custom cuts per jurisdiction, language, date range or document type are prepared within 5–6 business days.
Ask for a cut -
Can we test it first?
Yes. Download the free sample now (1,233 records · 54 document histories (114 versions) · 2,017 typed edges, including the GDPR, AI Act, Digital Markets Act and Digital Services Act), or read real rows in the record explorer above. CJEU case law: an excerpt under an evaluation licence on request. Then ask for a cut on your own jurisdictions.
Download the free sample
Licensing
Annual licences, priced in public.
Commercial and Foundation / Lab include 12 months of monthly updates and a perpetual right to the snapshots you received.
Open sample
Evaluating fit, academic work, demos.
Free
upstream licences per record
- 1,233 records · 54 document histories (114 versions) · 2,017 typed edges, same schema as the licensed snapshot
- Includes the GDPR, AI Act, Digital Markets Act and Digital Services Act
- Each record carries its own upstream licence and attribution
Research / Startup
Companies under about 50 staff, research labs.
from €7,000
per year
- Corpus, history and graph for the lanes you choose
- Internal R&D, evals and fine-tuning
- RAG behind your product for internal use and pilots
Commercial
productionLegal-AI products in production.
from €18,000
per year
- Everything in Research
- Production RAG and fine-tuned models in your hosted services
- 12 months of monthly updates
Foundation / Lab
Foundation-model developers and AI labs.
from €50,000
per year · custom terms
- Everything in Commercial
- Pre-training and distribution in model weights
- Custom exports, cuts and delivery
Rights matrix, update terms and tombstone obligations →
Licence agreement on request: we send the master licence and your tier schedule with the sample, and quote for your lanes. We reply within one business day.
For your procurement team
- Contracting entity
- Arieta Labs Ltd, England & Wales, company no. 16629785
- Governing law
- England and Wales
- Data protection
- Controller-to-controller data-protection terms in the licence agreement, sent on request
- Delivery
- Parquet (zstd) and gzip JSONL with MANIFEST.json (SHA-256 per shard), DATASHEET.md and LICENCES.md; signed download or your bucket, within 5–6 business days
- Updates
- Monthly deltas; upstream withdrawals are reflected as tombstones in the next delta
- Personal data
- No raw upstream HTML; national case law delivered with personal data masked
- Website
- No cookies; cookie-free Cloudflare Web Analytics only
- Certifications
- None claimed
Who is behind it
A real company, an open method, no made-up logos.
Arieta Data is the dataset business of Arieta Labs Ltd. Check everything below before you talk to us.
-
Company
Arieta Labs Ltd, London. Company no. 16629785, England & Wales.
-
Same corpus, live API
The corpus also powers LexAPI, a REST API and MCP server over EU and national law.
-
Founder-led
Nejc Fošnarič builds the pipeline behind the data and writes the guides on this site.
-
Open methodology
Datasheets, schema, Croissant metadata and the counting method are public.
-
Open source
The LexAPI SDKs and MCP server are on GitHub.
Documents and citation edges are exact counts. Articles and tokens are estimated from a 2% random sample of rows, extrapolated per lane (small lanes are counted exactly); tokens are characters ÷ 4.
AI Act · Article 53(1)(d)
Your training-data summary, already drafted.
Providers of general-purpose AI models must publish a summary of their training content (Regulation (EU) 2024/1689, applicable since 2 August 2025). Here is the paragraph for this dataset, with this snapshot’s numbers.
Licensed dataset: Arieta Data, from Arieta Labs Ltd (England & Wales company no. 16629785), snapshot 2026-11.
Data type: text (legislation and court decisions), structured at article level, with metadata, version history and citation edges.
Size: 419,355 documents, approximately 4,070,000 articles, 15,098 document versions, 801,149 citation edges, approximately 2,050,000,000 tokens (characters/4, estimated from a sample), as of the November 2026 snapshot.
Languages: English, French, Lithuanian, Croatian, Dutch, Polish, German, Danish, Slovenian, Czech, Spanish, Swedish, Finnish, Italian, Bulgarian, Romanian, Maltese, Hungarian, Portuguese, Estonian, Greek, Latvian, Slovak, Irish.
Time span of content: 1801-06-10 to 2026-11-12.
Source category: private dataset obtained from a third party (Arieta Labs Ltd), compiled from publicly available official publications:
- European Union (EU): EUR-Lex legal notice (reuse authorised, source acknowledged; Commission documents under Decision 2011/833/EU)
- Austria (AT): CC BY 4.0 (RIS OGD, Namensnennung 4.0 International)
- Czech Republic (CZ): NKOD podmínky užití: no copyright work, no database right, no sui generis right (≈ CC0)
- Germany (DE): Official works outside copyright (§ 5(1) UrhG); no licence stated by GII / RII
- Denmark (DK): Betingelser for anvendelse af Civilstyrelsens åbne data (gratis, ikke-eksklusiv og ubegrænset brugsret; ophavsretslovens § 9)
- Spain (ES): Condiciones de reutilización de la AEBOE (licencia tipo, Resolución de 27 de junio de 2024; Ley 37/2007)
- Finland (FI): Creative Commons Attribution 4.0 International (CC BY 4.0) — Finlex open data
- France (FR): Licence Ouverte 2.0 (+ Judilibre CGU for case law)
- Croatia (HR): Official texts outside copyright (ZAPSP NN 111/2021 čl. 18 st. 3); odluke.sudovi.hr Uvjeti korištenja (automated download, any use, attribution)
- Italy (IT): CC BY 4.0 (Normattiva open data, in force from 1 Jan 2026; all records fetched in 2026; OpenGA: CC BY 4.0)
- Lithuania (LT): Creative Commons Attribution 4.0 (data.gov.lt datasets 2613 and 1938); legal acts and court decisions outside copyright (Autorių teisių įstatymas, art. 5)
- Netherlands (NL): No copyright on statutes and judgments (Auteurswet art. 11) and no database right (Databankenwet art. 8); Rechtspraak open data Public Domain Mark / CC0
- Poland (PL): No licence stated; official acts and judgments excluded from copyright (art. 4 Prawo autorskie), re-use under the Open Data Act 2021
- Sweden (SE): Riksdagens öppna data: free use and redistribution with attribution ("Källa: Sveriges riksdag"); court decisions and statutes exempt from copyright (9 § URL 1960:729)
- Slovenia (SI): CC BY 4.0 (Priznanje avtorstva), OPSI records for PISRS NPB and Sodna praksa
- United Kingdom (UK): Open Government Licence v3.0
Collection: Arieta Labs collected the upstream data through publishers' documented APIs, open-data dumps and official websites. Documents that a source lists for removal (for example takedowns listed in robots.txt) and documents withdrawn by the source are removed and delivered as tombstones. A per-source record of terms and access channels is supplied with each snapshot.
Personal data: national court decisions are not in this snapshot’s files; they are delivered separately, only after personal-data masking. CJEU case law is included as published by the Court. Documents withdrawn by their source are delivered as tombstones and deleted by the licensee.
Copyright and reuse: each record carries its source, licence and attribution. Reuse of EUR-Lex content is permitted under the EUR-Lex legal notice (Commission reuse policy, Decision 2011/833/EU); CJEU case law under the CURIA copyright notice; national texts under the terms listed per source. UK case law is not included: the Open Justice Licence does not cover computational analysis.
More: https://data.arietalabs.com/provenance/ FAQ
Questions buyers ask first.
What is Arieta Data?
Arieta Data is a licensable dataset of European law for AI training, retrieval and evaluation, from Arieta Labs Ltd, the company behind LexAPI. It contains EU legislation, CJEU case law and national legislation from 15 countries, parsed to article level, with version history and a typed citation graph. As of the November 2026 snapshot the corpus holds 3.9M+ documents, of which 410k+ documents (4M+ articles) are ready to license.
Can I legally train an AI model on EUR-Lex data?
EUR-Lex’s legal notice allows reuse of the legal documents it publishes, including commercially, with acknowledgement of the source; Commission documents follow the Commission’s reuse policy (Decision 2011/833/EU) and CJEU case law the CURIA copyright notice. National sources each have their own terms, and UK case law needs a separate licence for computational analysis. Arieta Data documents the terms per source. See Is EUR-Lex data licensable for AI training?. Not legal advice.
Is national case law included?
Yes. CJEU case law is included as published by the Court. National case law is included too (3.4M+ decisions as of the November 2026 snapshot): court decisions name or pseudonymise people, so they are delivered with personal data masked, including magistrate and clerk names for France. Tell us which courts you need, subject to licence terms. UK case law is not offered: the Open Justice Licence does not cover computational analysis, and the separate licence from The National Archives cannot be passed on.
What formats, delivery and updates do you offer?
Apache Parquet (zstd, sharded at about 512 MB) and gzip JSON Lines with the same fields, by signed download or to a bucket you control. Each snapshot ships MANIFEST.json with SHA-256 checksums per shard, a datasheet and LICENCES.md. Updates are monthly deltas with upserts and tombstones.
How fast can we get the data?
The free sample downloads now, with no sign-up. Custom cuts per jurisdiction, language, date range or document type, and licensed deliveries, are prepared within 5–6 business days. We reply within one business day. See Request access.
How much does it cost?
Sample records are free to download. Research / Startup licences start at €7,000 a year, Commercial at €18,000 a year, and Foundation / Lab at €50,000 a year (custom terms). Commercial and Foundation / Lab include a perpetual right to the snapshots you received. See Licensing.
See the rows for your use case.
Tell us your jurisdictions, languages and use case. The free sample downloads today; a custom cut by jurisdiction, language, date range or document type, with the licence agreement, is ready within 5–6 business days.
- We reply within one business day
- Custom cuts in 5–6 business days
- Free sample, no sign-up