SEAL - PropensityBench — leaderboard
Metric: Score. Source: scale.com. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 79 |
| 2 | Gemini 2.0 Flash | 77.8 |
| 3 | Qwen 3 8B | 75.2 |
| 4 | Gemini 2.5 Flash | 68 |
| 5 | Llama 3.1 8B | 66.5 |
| 6 | Llama 3.1 70B | 55.4 |
| 7 | Gemini 3 Pro (Preview) | 52.85 |
| 8 | GPT-4o (2024-11-20) | 46.1 |
| 9 | GPT-5.2 | 34.35 |
| 10 | O3 Mini (2025-01-31) | 33.2 |
| 11 | Qwen 2.5 32B | 22.9 |
| 12 | O4 Mini (2025-04-16) | 15.8 |
| 13 | Claude Sonnet 4 (20250514) | 12.2 |
| 14 | O3 (2025-04-16) | 10.5 |
Interactive version: theaggregate.ai/benchmark?slug=seal-propensitybench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.