APEX v1 Big Law — leaderboard
Metric: Score (%). Source: www.mercor.com. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (High) | 76.6 |
| 2 | Claude Opus 4.6 (High) | 76.4 |
| 3 | Claude Opus 4.6 (Max) | 76.2 |
| 4 | GPT-5.4 (High) | 74.3 |
| 5 | O3 (High) | 73.6 |
| 6 | GPT-5 Codex (High) | 73.5 |
| 7 | GPT-5.2 (High) | 73.4 |
| 8 | GPT-5.1 Codex (High) | 73.2 |
| 9 | GPT-5.2 Codex (High) | 73 |
| 10 | Claude Sonnet 4.6 (High) | 72.3 |
| 11 | Claude Opus 4.5 (High) | 70.1 |
| 12 | Grok 4.1 Fast | 70 |
| 13 | GPT-5.3 Codex (High) | 69.8 |
| 14 | Gemini 3.1 Pro (Preview) (High) | 67.6 |
| 15 | Grok 4 | 67 |
Interactive version: theaggregate.ai/benchmark?slug=apex-v1-big-law · How the rankings work · Data refreshed daily, snapshot 2026-07-22.