LexRubric - Completeness: leaderboard

Metric: Dimension score rate (%): points earned from expert-written atomic rubric items of the completeness 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
1Kimi K2.673.85
2Qwen 3.6 Max Preview72.02
3Kimi K2.570.91
4Qwen 3 Max70.84
5Claude Sonnet 4.669.71
6GLM-5.165.15
7Qwen 3.5 397B A17B64.95
8GPT-5.263.94
9DeepSeek V4 Flash61.11
10DeepSeek R157.2
11GLM-554.98
12DeepSeek V4 Pro54.1
13DeepSeek V3.252.93

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