Data Extraction (Applications)
Systems extract requested information from legal documents. The benchmark checks whether the resulting work satisfies the task criteria.
The publisher reports results for applications under its track methodology.
- Made by
- Legal Benchmarks
- Kind
- Law
- Level
- Expert
- Scored
- Graded by people
- Answers
- Open-ended
- Measures
- Capability
- Subject
- Law
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- A host
- Based on
- A version of another benchmark
- In views
- Law and finance
DeepSeek Harness (V4 Pro) tops the leaderboard with 48.3%
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
- 48.3 DeepSeek Harness (V4 Pro)
- 43.3 Claude Web App (Fable 5)
- 38.3 ChatGPT Codex (GPT 5.6 Sol)
- 38.3 ChatGPT Web App (GPT 5.6 Sol)
- 38.3 Claude Cowork (Fable 5)
- 38.3 Claude Web App (Sonnet 5)
- 38.3 Grok Web App (4.6)
- 33.3 NotebookLM
- 30.0 Gemini Web App (Gemini 3.1 Pro)
- 30.0 Perplexity Computer
- 20.0 Microsoft Copilot Web App
- 1 DeepSeek Harness (V4 Pro) 48.3%
- 2 Claude Web App (Fable 5) 43.3%
- 3 ChatGPT Codex (GPT 5.6 Sol) 38.3%
- 4 ChatGPT Web App (GPT 5.6 Sol) 38.3%
- 5 Claude Cowork (Fable 5) 38.3%
- 6 Claude Web App (Sonnet 5) 38.3%
- 7 Grok Web App (4.6) 38.3%
- 8 NotebookLM 33.3%
- 9 Gemini Web App (Gemini 3.1 Pro) 30%
- 10 Perplexity Computer 30%
- 11 Microsoft Copilot Web App 20%
Source: Official Data Extraction (Applications) leaderboard
| # | Entry | Score | Setting | Agent by | Date | Source |
|---|---|---|---|---|---|---|
| 1 | DeepSeek Harness (V4 Pro) | 48.3% | DeepSeek Harness (V4 Pro) | DeepSeek | Official leaderboard | |
| 2 | Claude Web App (Fable 5) | 43.3% | Claude Web App (Fable 5) | Anthropic | Official leaderboard | |
| 3 | ChatGPT Codex (GPT 5.6 Sol) | 38.3% | ChatGPT Codex (GPT 5.6 Sol) | OpenAI | Official leaderboard | |
| 4 | ChatGPT Web App (GPT 5.6 Sol) | 38.3% | ChatGPT Web App (GPT 5.6 Sol) | OpenAI | Official leaderboard | |
| 5 | Claude Cowork (Fable 5) | 38.3% | Claude Cowork (Fable 5) | Anthropic | Official leaderboard | |
| 6 | Claude Web App (Sonnet 5) | 38.3% | Claude Web App (Sonnet 5) | Anthropic | Official leaderboard | |
| 7 | Grok Web App (4.6) | 38.3% | Grok Web App (4.6) | xAI | Official leaderboard | |
| 8 | NotebookLM | 33.3% | NotebookLM | NotebookLM | Official leaderboard | |
| 9 | Gemini Web App (Gemini 3.1 Pro) | 30% | Gemini Web App (Gemini 3.1 Pro) | Official leaderboard | ||
| 10 | Perplexity Computer | 30% | Perplexity Computer | Perplexity | Official leaderboard | |
| 11 | Microsoft Copilot Web App | 20% | Microsoft Copilot Web App | Microsoft | Official leaderboard |
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Law, finance and medicine
- MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy.
Qwen 3.8 Max 80.4%
- HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
- HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy.
Ling-3.1-flash 81.7%
- KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
- Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%
Data audit
7 of 8 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (11 entries).
- Pass Models scored 11 entries on its leaderboard, none a model we track.
- Note Newest score No scores yet.
- Note Frontier models No models scored yet.
- Note Headroom No scores yet.
Leaderboard import 3 pass
- Pass Importer We read the board on legalbenchmarks.ai: 11 entries at the last read, none a model we track.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.