PolicyBench - US Exact Match: leaderboard
United States household tax and benefit calculations without tools. The headline household-impact-weighted exact-match metric allows a one-dollar numeric tolerance and exact binary coverage decisions; households receive equal final weight. Native per-model scored-output counts and symmetric reference exclusions are retained. Forced answer-tool calls disable thinking for affected Claude configurations; JSON-contract rows retain their native reasoning settings. Logged run-state settings remain distinct from registry-derived settings. Continuous score, bounded score, cost and latency columns are separate measurements. Metric: Household-impact-weighted exact match (%). Source: https://policybench.org
Metric: Household-impact-weighted 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 | 95.16 | #29 |
| 2 | Claude Opus 5.5 | 94.07 | #7 |
| 3 | GPT-6.1 Terra | 93.14 | #22 |
| 4 | GPT-5.6 Sol | 93.08 | #33 |
| 5 | Claude Fable 5.2 | 91.95 | #1 |
| 6 | GPT-6 Astra | 91.73 | #8 |
| 7 | Claude Sonnet 5.5 | 91.67 | #21 |
| 8 | GPT-6 Luna | 91.5 | #105 |
| 9 | Claude Fable 5.1 | 90.84 | #17 |
| 10 | Kimi K3 | 90.53 | #59 |
| 11 | GPT-6.1 Sol | 90.16 | #12 |
| 12 | Muse Wonder | 90.05 | #31 |
| 13 | GPT-5.6 Luna | 88.7 | #130 |
| 14 | Gemini 4 Argon | 88.61 | #14 |
| 15 | GLM-5.3 Flash | 88.57 | #136 |
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 · How It Works · Data refreshed daily, snapshot 2026-11-17.