SAD - Stages (Situating Prompt): leaderboard
Metric: Score (%, higher is better). Source: situational-awareness-dataset.org. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 Preview (2024-09-12) | 61.52 |
| 2 | O1 Mini (2024-09-12) | 52.73 |
| 3 | Claude 3 Opus | 51.09 |
| 4 | Claude 3.5 Sonnet | 49.31 |
| 5 | Llama 3 70B Chat | 48.44 |
| 6 | GPT-4 (0613) | 48.31 |
| 7 | GPT-4 Preview (0125) | 46.88 |
| 8 | Claude 3 Sonnet | 44.19 |
| 9 | GPT-4o | 44.06 |
| 10 | GPT-4 Base | 43.31 |
| 11 | Llama 2 70B Chat | 42.31 |
| 12 | Claude 2.1 | 41.5 |
| 13 | GPT-3.5 Turbo (0613) | 41.39 |
| 14 | Llama 2 13B Chat | 40.83 |
| 15 | Claude Instant 1.2 | 39.94 |
Interactive version: theaggregate.ai/benchmark?slug=sad-stages-situating-prompt · How It Works · Data refreshed daily, snapshot 2026-09-19.