Benchmark Almanac

Harvey LAB-AA v1.1

GPT-6.2 Astra 15% leads since November 2026
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.

Artificial Analysis says scores do not compare with version 1.0, whose headline was the criterion pass rate. The setup also differs from Harvey's own runs on the Harvey LAB record.

Released
October 2026
Kind
Law
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
Kept private
Measures
Capability
Subject
Law
Language
English
In and out
Text, Documents → Documents, Actions
Interaction
Agentic
Environment
Terminal
Items from
Another benchmark
Run by
A host
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last in a launch post
November 2026
Last model added
November 2026
5
Models scored in lab cards
4
Launch posts list it, from 2 labs
3
System cards report it, from 2 labs

GPT-6.2 Astra leads with 15%

Best score per model as labs report them in their own cards, drawn striped. Labs pick their own settings, so these compare less cleanly than independent runs.

  1. 1 GPT-6.2 Astra 15%
  2. 2 Muse Wonder Full 13.1%
  3. 3 GPT-6.2 Sol 11.1%
  4. 4 Muse Wonder 10%
  5. 5 Claude Fable 5.2 max effort 7.20%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 GPT-6.2 Astra 15% Nov 2026 Lab model card
2 Muse Wonder Full 13.1% Oct 2026 Lab model card
3 GPT-6.2 Sol 11.1% Nov 2026 Lab model card
4 Muse Wonder 10% Oct 2026 Lab model card
5 Claude Fable 5.2 7.20% max effort Nov 2026 Lab model card

2 models have held number one in labs’ own numbers since its release in October 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Meta OpenAI
Lab claims 2 leaders

Source: Labs’ model cards

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6.2 Astra 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6.2 Astra

Source: Labs’ model cards

2 labs put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out. Meta and OpenAI account for 4 of the 5 releases that named it since 2024.

Meta
OpenAI
DateLaunch postScoreAs the post writes it
Nov 17, 2026 GPT-6.2 Sol 11.1 Harvey-LAB-AA v1.1 (GPT-6.2 Sol Max)
Nov 17, 2026 GPT-6.2 Astra 15 Harvey-LAB-AA v1.1 (GPT-6.2 Astra Max)
Oct 12, 2026 Muse Wonder 10 Harvey LAB-AA
Oct 12, 2026 Muse Wonder Full 13.1 Harvey LAB-AA

3 system cards report it, from 2 labs

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Nov 9, 2026 Claude Fable 5.2 system card 7.2 Harvey LAB-AA v1.1, hallucination-gated all-pass rate, max effort, updated private set, three judges. All-pass before hallucination gate 10.8%. Not comparable with v1.0 or Harvey public LAB (Section 8.14.2) Capabilities
Oct 12, 2026 Muse Wonder model card 10 Harvey LAB-AA, Artificial Analysis version 1.1 with hallucination checks, Muse Wonder (max) (capabilities table image, page 3) Capabilities
Oct 12, 2026 Muse Wonder Full model card 13.1 Harvey LAB-AA, Artificial Analysis version 1.1 with hallucination checks, Muse Wonder Full (max) (capabilities table image, page 3) Capabilities

Versions

More in Law, finance and medicine

All 637 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%
  • KBL In use Models answer Korean legal knowledge and reasoning tasks and Korean bar exam questions. It has 7 knowledge tasks, 4 reasoning tasks and 2,510 bar exam items. Solar Pro 4 77.5%

Data audit

All 6 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Note Independent runs Only labs’ own numbers, as expected for an in-house eval. Scores from launch posts (4 scores) and system cards (3 scores).
  • Pass Models scored 5 models we track.
  • Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
  • Note Frontier models 2 of the top 10 have a score, which is fine for an in-house eval. The strongest tested is GPT-6.2 Astra (#1).
  • Pass Headroom The best is 15%, under 85%, so it still separates models.
Leaderboard import 1 note
  • Note Importer No public leaderboard to read.
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.

How we check