GPT-6.1 Terra
United States benchmark
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%
- Person-level WIC eligibility100.0%0.3%
- Reduced-price school meals eligibility100.0%0.1%
- Self-employment tax100.0%2.1%
- SSI100.0%2.0%
- TANF99.0%0.3%
- Payroll tax99.0%16%
- Person-level CHIP eligibility98.9%0.2%
- Person-level Medicaid eligibility97.7%30%
- Person-level Medicare eligibility97.1%11%
- Free school meals eligibility96.0%0.7%
- Federal refundable credits93.0%3.7%
- SNAP87.1%4.1%
- State refundable credits83.8%0.6%
- Federal tax before refundable credits83.3%19%
- State tax before refundable credits67.7%6.0%
Table view (exact, within 1%, outputs), hardest first
| Program | Exact | Within 1% | n |
|---|---|---|---|
| State tax before refundable credits | 67.7% | 68.8% | 93 |
| Federal tax before refundable credits | 83.3% | 85.7% | 84 |
| State refundable credits | 83.8% | 84.8% | 99 |
| SNAP | 87.1% | 90.3% | 93 |
| Federal refundable credits | 93.0% | 93.0% | 100 |
| Free school meals eligibility | 96.0% | 96.0% | 100 |
| Person-level Medicare eligibility | 97.1% | 97.1% | 172 |
| Person-level Medicaid eligibility | 97.7% | 97.7% | 173 |
| Person-level CHIP eligibility | 98.9% | 98.9% | 177 |
| Payroll tax | 99.0% | 100.0% | 96 |
| TANF | 99.0% | 99.0% | 100 |
| Local income tax | 100.0% | 100.0% | 98 |
| Person-level Early Head Start eligibility | 100.0% | 100.0% | 38 |
| Person-level Head Start eligibility | 100.0% | 100.0% | 38 |
| Person-level WIC eligibility | 100.0% | 100.0% | 177 |
| Reduced-price school meals eligibility | 100.0% | 100.0% | 100 |
| Self-employment tax | 100.0% | 100.0% | 100 |
| SSI | 100.0% | 100.0% | 97 |
Hardest casesworst misses on positive references
- Federal refundable credits250% offReference $822 · predicted $2,878Inspect household #100
- SNAP100% offReference $240 · predicted $0Inspect household #013
- State refundable credits100% offReference $25 · predicted $0Inspect household #013
- Federal tax before refundable credits100% offReference $643 · predicted $0Inspect household #023
- SNAP100% offReference $288 · predicted $0Inspect household #027
- State refundable credits100% offReference $62 · predicted $0Inspect household #028
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 9,050 rows whose legacy threshold score is below 1. That universe contains 9,046 of 9,046 exact-match misses and 4 exact hits. Another 2,407 rows with bounded score below 100 were not selected and have no audit annotation.