ReLE - Law and Civil Service: leaderboard
ReLE Chinese LLM benchmark (NoneLinear, formerly CLiB), a continuously updated Chinese-language evaluation run through each model's API. Law and civil service domain: the national lawyer qualification exam (including JEC-QA items) and civil service exams. Accuracy 0-100, higher is better.
Metric: Accuracy (%). Source: nonelinear.com. Saturation forecast: Around February 2027. 147 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 89 | #33 |
| 2 | GPT-6.1 Sol | 87.3 | #7 |
| 3 | Qwen 3.7 Plus | 87 | #98 |
| 4 | GLM-5.2 | 86.7 | #80 |
| 5 | Qwen 3.8 Flash | 86.3 | #62 |
| 6 | Qwen 3.8 Max | 86 | #34 |
| 7 | Grok 4.5 | 86 | #54 |
| 8 | GPT-6 Astra | 85.7 | #6 |
| 9 | Kimi K3 | 85.3 | #35 |
| 10 | DeepSeek V4.1 Flash | 85.3 | #37 |
| 11 | Qwen 3.7 Max | 85.3 | #85 |
| 12 | Hy4 preview | 85.3 | #45 |
| 13 | Kimi K3.1 Pro | 85 | #17 |
| 14 | Qwen 3.5 122B A10B | 84.7 | #184 |
| 15 | Claude Opus 5 | 84.7 | #16 |
No result here: #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=rele-law-and-civil-service · How It Works · Data refreshed daily, snapshot 2026-11-05.