PolicyBench - US Exact Match - snap: leaderboard

The snap output of PolicyBench US, without tools. Read countries.us.heatmap exact, with the native common per-output denominator and symmetric reference exclusions. Aggregate programStats rows across models are not model scores. Metric: Exact match (%). Source: https://policybench.org

Metric: Exact match (%). Source: policybench.org. Saturation forecast: Around February 2027. 52 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-6.1 Sol92.47#12
2GPT-6 Astra91.4#8
3Claude Opus 5.588.17#7
4GPT-6 Sol88.17#29
5GPT-5.6 Sol87.1#33
6Claude Fable 5.187.1#17
7GPT-6.1 Terra87.1#22
8Claude Fable 5.287.1#1
9GPT-5.586.02#50
10GLM-5.3 Flash86.02#136
11GPT-5.6 Terra86.02#68
12Muse Wonder86.02#31
13Kimi K384.95#59
14MiniMax-M384.95#278
15GPT-5.6 Luna84.95#130

No result here: #2 GPT-6.2 Astra, #3 Gemini 4 Carbon, #6 GPT-6.2 Sol.

Interactive version: theaggregate.ai/benchmark?slug=policybench-us-exact-match-snap · How It Works · Data refreshed daily, snapshot 2026-11-17.