Legal Research Bench: leaderboard
Legal Research Bench is a text benchmark for legal research tasks across diverse areas of US law. The evaluated model performs legal research, and the pinned metric is Score (%), measuring performance on those research tasks. Source: BenchmarkList’s attributed result table.
Metric: Score (%). Source: benchmarklist.com. Saturation forecast: Around February 2027. 27 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 55.29 | #16 |
| 2 | Claude Fable 5 | 49.52 | #20 |
| 3 | GPT-5.6 Sol | 48.08 | #22 |
| 4 | Kimi K3 | 44.23 | #35 |
| 5 | Claude Opus 4.8 | 43.75 | #39 |
| 6 | Claude Sonnet 5 | 41.83 | #65 |
| 7 | GPT-5.6 Terra | 40.87 | #40 |
| 8 | GPT-5.5 | 40.39 | #33 |
| 9 | Claude Sonnet 4.6 | 38.46 | #94 |
| 10 | Grok 4.5 | 37.98 | #54 |
| 11 | Muse Spark 1.1 | 37.98 | #55 |
| 12 | GPT-5.6 Luna | 36.54 | #79 |
| 13 | GLM-5.2 | 31.25 | #80 |
| 14 | Gemini 3.5 Flash | 30.77 | #66 |
| 15 | MiniMax-M3 | 29.81 | #151 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=legal-research-bench · How It Works · Data refreshed daily, snapshot 2026-11-05.