Harvey BigLaw Bench
- Tests
- Harvey benchmark context for BigLaw-style legal tasks; useful market signal but not an open runnable benchmark.
- Best for
- Legal AI market context
- Classification
- scoreboard · current · commercial
AI evaluation domain · 11 source-backed records
Legal AI Benchmarks
Legal benchmarks can reveal reasoning, retrieval, drafting, and document-review limits, but results are jurisdiction- and task-dependent. Use public sets to shortlist systems, then validate against your own authorities and workflows.
Decision guidance
How should legal teams use public AI benchmarks?
Check the jurisdiction, task definitions, source dates, and grading method before interpreting a score. Build private tests around citations, deadlines, forms, privilege, abstention, and the specific work product your professionals review.
11
Benchmarks in this domain
Sources first · alphabetical collection order
Runnability describes access to a usable repository, dataset, or harness—not whether setup is easy.
2026-07-15
Harvey Legal Agent Benchmark / LAB
- Tests
- Open legal-agent benchmark with long-horizon tasks across practice areas and expert rubric criteria.
- Best for
- Agentic legal work product
- Classification
- agent benchmark · current · open
LawBench
- Tests
- Legal tasks across entity recognition, reading comprehension, legal consultation, and more.
- Best for
- Chinese/legal reasoning research
- Classification
- benchmark · current · open
Legal RAG Bench
- Tests
- End-to-end legal retrieval and reasoning on realistic research tasks.
- Best for
- Comparing legal RAG systems where sources and reasoning both matter.
- Classification
- benchmark · emerging · open
LegalBench
- Tests
- 162 legal reasoning tasks contributed by lawyers, law professors, researchers, and legal practitioners.
- Best for
- Legal reasoning baseline
- Classification
- benchmark · current · open
LegalBench-RAG
- Tests
- Legal retrieval and generation benchmark for end-to-end legal RAG systems.
- Best for
- Legal document retrieval and grounded answers
- Classification
- benchmark · current · open
LEXam
- Tests
- Legal reasoning on hundreds of Swiss, EU, and international law examination questions.
- Best for
- Comparing legal knowledge across European and international jurisdictions.
- Classification
- benchmark · emerging · open
LexGLUE
- Tests
- Legal NLP benchmark suite in a SuperGLUE-like format.
- Best for
- Classic legal NLP tasks
- Classification
- benchmark · legacy · open
PLawBench
- Tests
- Legal consultation, case analysis, and document generation with workflow-grounded rubrics.
- Best for
- Evaluating legal-practice outputs rather than multiple-choice legal knowledge.
- Classification
- benchmark · emerging · open
RedlineBench
- Tests
- Multi-turn contract redlining and negotiation behavior.
- Best for
- Evaluating legal agents that must preserve objectives across revisions.
- Classification
- benchmark · emerging · open
Vals AI LegalBench
- Tests
- Hosted leaderboard for legal-model evaluation.
- Best for
- Current legal model comparison
- Classification
- scoreboard · current · commercial