PolicyBench - US Exact Match - ssi: leaderboard
The ssi 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 | GPT-5.6 Sol | 100 | #33 |
| 2 | GPT-6 Astra | 100 | #8 |
| 3 | Claude Opus 5.5 | 100 | #7 |
| 4 | GPT-6 Luna | 100 | #105 |
| 5 | GPT-6.1 Sol | 100 | #12 |
| 6 | Claude Sonnet 5.5 | 100 | #21 |
| 7 | Grok 4.7 | 100 | #48 |
| 8 | GPT-6.1 Terra | 100 | #22 |
| 9 | Muse Wonder | 100 | #31 |
| 10 | GLM-5.3 Flash | 98.97 | #136 |
| 11 | Claude Fable 5.1 | 98.97 | #17 |
| 12 | GPT-6 Sol | 98.97 | #29 |
| 13 | Inkling | 98.97 | #282 |
| 14 | Claude Fable 5.2 | 98.97 | #1 |
| 15 | Muse Spark 1.3 | 98.97 | #45 |
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-ssi · How It Works · Data refreshed daily, snapshot 2026-11-17.