SEAL - PropensityBench — leaderboard

Metric: Score. Source: scale.com. 14 models tracked.

Top models

#ModelScore
1Gemini 2.5 Pro79
2Gemini 2.0 Flash77.8
3Qwen 3 8B75.2
4Gemini 2.5 Flash68
5Llama 3.1 8B66.5
6Llama 3.1 70B55.4
7Gemini 3 Pro (Preview)52.85
8GPT-4o (2024-11-20)46.1
9GPT-5.234.35
10O3 Mini (2025-01-31)33.2
11Qwen 2.5 32B22.9
12O4 Mini (2025-04-16)15.8
13Claude Sonnet 4 (20250514)12.2
14O3 (2025-04-16)10.5

Interactive version: theaggregate.ai/benchmark?slug=seal-propensitybench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.