ActTraitBench — leaderboard

Metric: G_KD (Global Knowledge-Decision Gap) (self-reported). Source: benchmarklist.com. 15 models tracked.

Top models

#ModelScore
1MiniMax-M2.52.17
2Gemini 3.1 Pro (Preview)1.83
3GLM-51.79
4Qwen 3 32B0.85
5DeepSeek V3.10.75
6DeepSeek V4 Flash0.65
7DeepSeek V3.20.59
8DeepSeek V4 Pro0.57
9Claude Sonnet 4.60.51
10DeepSeek V30.47
11GPT-4o0.28
12Qwen 3 8B0.23

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