PolicyBench - US Exact Match - local_income_tax: leaderboard
The local_income_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: Estimated already saturated. 52 models tracked. Updated 17 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 100 | #99 |
| 2 | Claude Sonnet 4.6 | 100 | #173 |
| 3 | GPT-5.5 | 100 | #50 |
| 4 | DeepSeek V4 Pro | 100 | #198 |
| 5 | Claude Opus 4.7 | 100 | #82 |
| 6 | GPT-5.6 Sol | 100 | #33 |
| 7 | GPT-5.4 Mini | 100 | #401 |
| 8 | Claude Opus 4.8 | 100 | #62 |
| 9 | MiniMax-M3 | 100 | #278 |
| 10 | Gemini 3.8 Flash | 100 | #39 |
| 11 | Claude Opus 5 | 100 | #23 |
| 12 | GPT-6 Astra | 100 | #8 |
| 13 | GLM-5.3 Flash | 100 | #136 |
| 14 | DeepSeek V4.1 Flash | 100 | #69 |
| 15 | Claude Opus 5.5 | 100 | #7 |
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-local-income-tax · How It Works · Data refreshed daily, snapshot 2026-11-17.