SEAL - Korean — leaderboard
Metric: Score. Source: scale.com. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 Preview | 66.43 |
| 2 | Claude 3.7 Sonnet (20250219) | 64.93 |
| 3 | GPT-4o (2024-05-13) | 64.58 |
| 4 | GPT-4.5 (Preview) | 63.76 |
| 5 | GPT-4 Turbo (Preview) | 60.76 |
| 6 | GPT-4o (2024-08-06) | 59.93 |
| 7 | Claude 3.5 Sonnet (20240620) | 59.38 |
| 8 | Gemini 2.0 Flash | 55.6 |
| 9 | Claude 3 Sonnet | 54.17 |
| 10 | Claude 3 Opus | 52.78 |
| 11 | GPT-4o Mini | 51.74 |
| 12 | GPT-4 | 51.39 |
| 13 | Llama 3.1 405B Instruct | 50.35 |
| 14 | Mistral Large 2 (Jul) | 50.35 |
| 15 | Gemini 1.5 Pro (0514) | 40.42 |
Interactive version: theaggregate.ai/benchmark?slug=seal-korean · How the rankings work · Data refreshed daily, snapshot 2026-07-22.