PolicyBench - US Exact Match - payroll_tax: leaderboard
The payroll_tax 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
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Sol | 100 | #29 |
| 2 | GPT-6.1 Terra | 98.96 | #22 |
| 3 | Claude Opus 5.5 | 95.83 | #7 |
| 4 | GPT-5.6 Sol | 94.79 | #33 |
| 5 | Claude Sonnet 5.5 | 94.79 | #21 |
| 6 | Grok 4.7 | 93.75 | #48 |
| 7 | Kimi K3 | 92.71 | #59 |
| 8 | GPT-6 Luna | 92.71 | #105 |
| 9 | Claude Fable 5.1 | 91.67 | #17 |
| 10 | Muse Spark 1.3 | 91.67 | #45 |
| 11 | Muse Wonder | 91.67 | #31 |
| 12 | GLM-5.3 Flash | 90.63 | #136 |
| 13 | GPT-5.6 Luna | 90.63 | #130 |
| 14 | Gemini 4 Argon | 90.63 | #14 |
| 15 | Grok 4.6 | 90.63 | #47 |
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-payroll-tax · How It Works · Data refreshed daily, snapshot 2026-11-17.