LaborBench: leaderboard
Legal RAG and statutory reasoning benchmark over U.S. unemployment-insurance law across multiple states.
Metric: F1 (imported). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.5 Haiku | 68.8 | #568 |
| 2 | DeepSeek V3 | 67.4 | #321 |
| 3 | GPT-4o Mini | 67 | #599 |
| 4 | Llama 3.1 70B Instruct | 64.7 | #561 |
| 5 | Gemini 1.5 Flash | 64.4 | #618 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=laborbench · How It Works · Data refreshed daily, snapshot 2026-10-20.