SEAL - Math — leaderboard
Metric: Score. Source: scale.com. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 96.6 |
| 2 | GPT-4o (2024-08-06) | 95.68 |
| 3 | Llama 3.1 405B Instruct | 95.6 |
| 4 | Claude 3 Opus | 95.19 |
| 5 | GPT-4 Turbo (Preview) | 95.1 |
| 6 | GPT-4o (2024-05-13) | 94.85 |
| 7 | Mistral Large 2 (Jul) | 93.94 |
| 8 | Claude 3 Sonnet | 93.28 |
| 9 | Gemini 1.5 Pro (0514) | 92.28 |
| 10 | Gemini 1.5 Flash | 90.12 |
| 11 | Llama 3 70B Instruct | 90.12 |
| 12 | Mistral Large | 87.47 |
| 13 | Gemini 1.0 Pro | 79.83 |
| 14 | CodeLlama-34B-Instruct | 37.51 |
Interactive version: theaggregate.ai/benchmark?slug=seal-math · How the rankings work · Data refreshed daily, snapshot 2026-07-22.