SEAL - Instruction Following — leaderboard
Metric: Score. Source: scale.com. 19 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 (2024-12-17) | 91.96 |
| 2 | DeepSeek R1 | 87.75 |
| 3 | Gemini 2.0 Flash | 86.58 |
| 4 | O1 Preview | 86.58 |
| 5 | Claude 3.5 Sonnet (20240620) | 85.96 |
| 6 | GPT-4o (2024-05-13) | 85.29 |
| 7 | Llama 3.1 405B Instruct | 84.85 |
| 8 | GPT-4 Turbo (Preview) | 83.19 |
| 9 | Mistral Large 2 (Jul) | 82.81 |
| 10 | GPT-4o | 82.52 |
| 11 | DeepSeek V3 | 82.34 |
| 12 | Llama 3.2 90B Vision Instruct | 82.07 |
| 13 | Llama 3 70B Instruct | 81.17 |
| 14 | GPT-4o (2024-08-06) | 80.17 |
| 15 | Claude 3 Opus | 80.12 |
Interactive version: theaggregate.ai/benchmark?slug=seal-instruction-following · How the rankings work · Data refreshed daily, snapshot 2026-07-22.