Benchmarks & Datasets

A curated list of awesome LegalTech resources - open source platforms, AI models, MCP servers, companies, datasets, and tools for the global legal ecosystem.

Open, structured corpus of US primary law: 2M+ sections of state statutory codes, the US Code, and state constitutions, plus the ingestion pipeline that builds it. Data on Hugging Face (CC BY 4.0); scrapers Apache-2.0. Quarterly snapshots.

Open, reproducible benchmark of US legal-answer quality. Verified questions, a standard-library scorer, and results anyone can rerun.

A benchmark built to evaluate and improve agent capabilities for supporting legal work.

Continually updated LLM and law papers

Open-source governance infrastructure for AI-assisted legal work

Benchmarking Legal Knowledge of Large Language Models

Machine-readable Italian legal corpus

AI-powered contract review skill with CUAD risk detection, market benchmarks, and lawyer-ready redlines. Works with Claude Code, Codex, Cursor, and 26+ tools.

We curate a large corpus of legal and administrative data. The utility of this data is twofold: (1) to aggregate legal and administrative data sources that demonstrate different norms and legal standards for data filtering; (2) to collect a dataset that can be used in the future for pretraining legal-domain language models, a key direction in access-to-justice initiatives.

Private benchmark ranking LLMs on legal

The code used to create and update the Open Australian Legal Corpus, the first and only multijurisdictional open corpus of Australian legislative and judicial documents.

LexEval: A Comprehensive Benchmark for Evaluating Large Language Models in Legal Domain

Open Australian Legal Corpus ‍⚖️ The Open Australian Legal Corpus by Isaacus, a foundational legal AI research company, is the first and only multijurisdictional open corpus of Australian legislative and judicial documents. Comprised of 229,122 texts totalling over 60 million lines and 1.4 billion tokens, the Corpus includes every in force statute and regulation in the Commonwealth, New South Wales, Queensland, Western Australia, South Australia, Tasmania and Norfolk Island, in… See the full description on the dataset page: https://huggingface.co/datasets/isaacus/open-australian-legal-corpus.

The Harvard USPTO Patent Dataset

LexGLUE: benchmark for legal language understanding

Multi Legal Pile is a dataset of legal documents in the 24 EU languages.

The EUR-Lex-Sum dataset is a multilingual resource intended for text summarization in the legal domain. It is based on human-written summaries of legal acts issued by the European Union. It distinguishes itself by introducing a smaller set of high-quality human-written samples, each of which have much longer references (and summaries!) than comparable datasets. Additionally, the underlying legal acts provide a challenging domain-specific application to legal texts, which are so far underrepresented in non-English languages. For each legal act, the sample can be available in up to 24 languages (the officially recognized languages in the European Union); the validation and test samples consist entirely of samples available in all languages, and are aligned across all languages at the paragraph level.

Multilingual EU law classification dataset

The Harvard USPTO Patent Dataset (HUPD) is a large-scale, well-structured, and multi-purpose corpus of English-language patent applications filed to the United States Patent and Trademark Office (USPTO) between 2004 and 2018. With more than 4.5 million patent documents, HUPD is two to three times larger than comparable corpora. Unlike other NLP patent datasets, HUPD contains the inventor-submitted versions of patent applications, not the final versions of granted patents, allowing us to study patentability at the time of filing using NLP methods for the first time.