LexKairos - Effective Date Inquiry: leaderboard
Metric: Accuracy (%; 485 items: generate the effective date of a named Chinese statute; vanilla setting, the model answers directly without extra reasoning steps (thinking disabled where the model allows it); temperature 0, 8,000 output tokens). Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Minimal) | 87.84 |
| 2 | DeepSeek V4 Flash (Non-reasoning) | 83.51 |
| 3 | GPT-5.4 (Non-reasoning) | 78.56 |
| 4 | DeepSeek R1 | 77.73 |
| 5 | Qwen 3.5 9B (Non-reasoning) | 44.95 |
| 6 | Llama 3.1 8B | 6.6 |
Interactive version: theaggregate.ai/benchmark?slug=lexkairos-effective-date-inquiry · How It Works · Data refreshed daily, snapshot 2026-09-29.