Skip to main content
PolicyBench

Claude Opus 4.7

Anthropic · AI alone, no tools
United States rank
#7 of 23
US exact
77.4%
United States

United States benchmark

Exact match
77.4%
Bounded score
88.7%
Parse rate
100.0%
Eligibility flags
905/984

Score by programhardest first

ProgramExactWithin 1%n
Federal tax before refundable credits54.0%56.0%100
State tax before refundable credits57.0%61.0%100
SNAP73.0%73.0%100
State refundable credits75.0%75.0%100
Federal refundable credits81.0%82.0%100
Payroll tax83.0%85.0%100
Person-level Medicaid eligibility86.4%86.4%177
Person-level Head Start eligibility86.8%86.8%38
Person-level CHIP eligibility88.1%88.1%177
Person-level Medicare eligibility89.8%89.8%177
Person-level WIC eligibility95.5%95.5%177
SSI96.0%96.0%100
Reduced-price school meals eligibility98.0%98.0%100
Free school meals eligibility99.0%99.0%100
TANF99.0%99.0%100
Local income tax100.0%100.0%100
Person-level Early Head Start eligibility100.0%100.0%38
Self-employment tax100.0%100.0%100

Hardest casesworst misses on positive references

Scores are from the frozen manuscript snapshot under the AI-alone condition: one structured response per household, no tools, graded against PolicyEngine reference outputs. See the leaderboard and paper for methodology.