Claude Opus 5
United States benchmark
Serving sensitivity This row ran under the board's forced answer-tool call, which switches Claude's extended thinking off. The same model, re-run with tool_choice: auto, scores 86.2% exact (+5.9) and would rank #5 on the live board; the board keeps the request shape its model card records. Read the note.
Exact match by programall 18 output groups, best first
Bar length is the unweighted exact-match rate on that program's outputs; the right column is the program's share of the headline weight, which follows household dollars, so the bars do not average to the headline.
- Local income tax100.0%0.3%
- Person-level Early Head Start eligibility100.0%3.1%
- Person-level Head Start eligibility100.0%1.2%
- TANF99.0%0.3%
- Person-level WIC eligibility98.9%0.3%
- Person-level CHIP eligibility98.3%0.2%
- Free school meals eligibility98.0%0.7%
- Reduced-price school meals eligibility98.0%0.1%
- Self-employment tax98.0%2.1%
- SSI96.9%2.0%
- Person-level Medicare eligibility96.5%11%
- Person-level Medicaid eligibility93.2%30%
- Payroll tax85.0%16%
- Federal refundable credits83.0%3.7%
- State refundable credits79.0%0.6%
- SNAP75.3%4.1%
- State tax before refundable credits59.0%6.0%
- Federal tax before refundable credits50.0%19%
Table view (exact, within 1%, outputs), hardest first
| Program | Exact | Within 1% | n |
|---|---|---|---|
| Federal tax before refundable credits | 50.0% | 50.0% | 100 |
| State tax before refundable credits | 59.0% | 61.0% | 100 |
| SNAP | 75.3% | 75.3% | 97 |
| State refundable credits | 79.0% | 80.0% | 100 |
| Federal refundable credits | 83.0% | 84.0% | 100 |
| Payroll tax | 85.0% | 90.0% | 100 |
| Person-level Medicaid eligibility | 93.2% | 93.2% | 177 |
| Person-level Medicare eligibility | 96.5% | 96.5% | 172 |
| SSI | 96.9% | 96.9% | 97 |
| Free school meals eligibility | 98.0% | 98.0% | 100 |
| Reduced-price school meals eligibility | 98.0% | 98.0% | 100 |
| Self-employment tax | 98.0% | 98.0% | 100 |
| Person-level CHIP eligibility | 98.3% | 98.3% | 177 |
| Person-level WIC eligibility | 98.9% | 98.9% | 177 |
| TANF | 99.0% | 99.0% | 100 |
| Local income tax | 100.0% | 100.0% | 100 |
| Person-level Early Head Start eligibility | 100.0% | 100.0% | 38 |
| Person-level Head Start eligibility | 100.0% | 100.0% | 38 |
Hardest casesworst misses on positive references
- State refundable credits1084% offReference $19 · predicted $225Inspect household #043
- SNAP1047% offReference $288 · predicted $3,300Inspect household #112
- Federal tax before refundable credits894% offReference $500 · predicted $4,966Inspect household #046
- SNAP793% offReference $288 · predicted $2,568Inspect household #073
- SNAP459% offReference $288 · predicted $1,608Inspect household #030
- Federal tax before refundable credits362% offReference $1,346 · predicted $6,210Inspect household #039
Scores are from the frozen manuscript snapshot under the AI-alone condition: structured responses over the same household facts (whole-scenario or in output subsets, per the serving-configuration table), no tools, graded against PolicyEngine reference outputs. See the leaderboard and paper for methodology. The frozen US annotations cover 8,783 rows whose legacy threshold score is below 1. That universe contains 8,780 of 8,780 exact-match misses and 3 exact hits. Another 1,605 rows with bounded score below 100 were not selected and have no audit annotation.