Benchmark Almanac

LongJud-Bench

Kimi K2.5 0.66 leads since June 2026
Not in model cards

Agents conduct connected stages of Chinese civil litigation in LegalWorld. The authors compare their legal consultations, court documents, and courtroom advocacy.

Made by
Songhan Zuo and colleagues
Released
June 2026
Kind
Law
Level
Expert
Scored
Graded by a model
Answers
Agent task
Tasks
All public
Measures
Capability
Subject
Law
Language
Chinese
In and out
Text → Text, Actions
Interaction
Multi-agent
Environment
LegalWorld, APIs and tools
Refresh
One fixed set
Run by
Anyone
Occupations
Legal
Strongest model tested
GPT-5.2 #51 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
6
Models scored in its paper

Kimi K2.5 leads with 0.66 among models released by April 2026

Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The cross-model table reports no overall score, so this is the plain mean of eight capabilities, averaging the two instances where applicable. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Kimi K2.5 0.66
  2. 2 Qwen3.5-Plus 0.65
  3. 3 GPT-5.2 0.63
  4. 4 DeepSeek-V4-Flash 0.63
  5. 5 GLM-4.7 0.61
  6. 6 Qwen3.5-Flash 0.55

Source: The LongJud-Bench paper

#ModelScoreSettingDateSource
1 Kimi K2.5 0.66 LegalWorld Jan 2026 Official leaderboard
2 Qwen3.5-Plus 0.65 LegalWorld Feb 2026 Official leaderboard
3 GPT-5.2 0.63 LegalWorld Dec 2025 Official leaderboard
4 DeepSeek-V4-Flash 0.63 LegalWorld Apr 2026 Official leaderboard
5 GLM-4.7 0.61 LegalWorld Dec 2025 Official leaderboard
6 Qwen3.5-Flash 0.55 LegalWorld Feb 2026 Official leaderboard

Rankings by part

We read 22 parts from the benchmark’s paper in 5 groups: legal consultation, document drafting, first instance, second instance and courtroom advocacy. Each ranks models on one part of the test. Kimi K2.5 leads 13 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Issue Spotting

6 models we track have a score here.

  1. 1 Kimi K2.5 0.67
  2. 2 GPT-5.2 ↑1 0.63
  3. 3 Qwen3.5-Plus ↓1 0.62
  4. 4 DeepSeek-V4-Flash 0.62
  5. 5 Qwen3.5-Flash ↑1 0.56
  6. 6 GLM-4.7 ↓1 0.54

Number one over time

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

Moonshot AI
Benchmark paper Not tested since Jun 2026
Kimi K2.5 0.66

Source: The LongJud-Bench paper

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
In the benchmark’s paper
0.55 0.60 0.65 2026 0.55 0.60 0.65 2026

Source: Official LongJud-Bench 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

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%
  • 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

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (6 entries).
  • Pass Models scored 6 models we track.
  • Warning Newest score The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.2 (#51).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 6 rows for models we track at the last read.
  • 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.

How we check