LexRubric - Legal Accuracy: leaderboard

Metric: Dimension score rate (%): points earned from expert-written atomic rubric items of the legal accuracy dimension (weighted -10 to +10; negative items subtract) divided by the maximum positive score, averaged over instances; a Qwen3.6-27B judge checks each item, temperature 0.6, 16k output tokens; can fall below 0; higher is better (instances without a positive item in the dimension are left out). Source: arxiv.org. Saturation forecast: Around August 2027. 18 models tracked.

Top models

#ModelScore
1Qwen 3.6 Max Preview69.51
2Kimi K2.668.53
3Qwen 3 Max68.45
4Kimi K2.566.6
5GLM-5.162.13
6Claude Sonnet 4.660.6
7Qwen 3.5 397B A17B59.46
8GLM-556.92
9DeepSeek V4 Flash55.34
10DeepSeek V4 Pro54.21
11DeepSeek V3.254.17
12DeepSeek R152.97
13GPT-5.237.75

Interactive version: theaggregate.ai/benchmark?slug=lexrubric-legal-accuracy · How It Works · Data refreshed daily, snapshot 2026-09-29.