PolicyBench - US Exact Match - person_head_start_eligible: leaderboard

The person_head_start_eligible 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: Estimated already saturated. 52 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-5.5100#50
2GPT-5.6 Sol100#33
3Gemini 3 Flash (Preview)100#151
4Gemini 3.8 Flash100#39
5Claude Opus 5100#23
6Gemini 3.5 Flash100#107
7GPT-6 Astra100#8
8GLM-5.3 Flash100#136
9DeepSeek V4.1 Flash100#69
10Claude Opus 5.5100#7
11GPT-5.6 Luna100#130
12GPT-5.6 Terra100#68
13Gemini 3.7 Flash100#38
14GPT-6 Luna100#105
15Gemini 3.6 Flash100#79

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-person-head-start-eligible · How It Works · Data refreshed daily, snapshot 2026-11-17.