Vals AI CaseLaw v2: leaderboard
Metric: Accuracy (%). Source: www.vals.ai. 54 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok 4.3 | 79.31 |
| 2 | GPT-5.1 (High) | 73.42 |
| 3 | GPT-4.1 | 69.88 |
| 4 | GPT-5 Mini (High) | 68.49 |
| 5 | Claude Opus 4.7 | 68.38 |
| 6 | GPT-5 (High) | 66.45 |
| 7 | GPT-5.5 (xHigh) | 66.24 |
| 8 | GPT-5.2 (xHigh) | 66.02 |
| 9 | Grok 4 | 65.81 |
| 10 | Kimi K2 (Thinking) | 65.7 |
| 11 | Grok 4 Fast (Reasoning) | 65.7 |
| 12 | Gemini 3.1 Pro (Preview) (High) | 64.84 |
| 13 | Command A | 64.52 |
| 14 | Claude Sonnet 4.6 | 63.99 |
| 15 | Gemini 2.5 Pro | 63.88 |
Interactive version: theaggregate.ai/benchmark?slug=vals-ai-caselaw-v2 · How It Works · Data refreshed daily, snapshot 2026-09-05.