Skip to main content
PolicyBench

Grok 4.7

xAI · AI alone, no tools
Current board, snapshot 2026-09-29
United States rank
#11 of 45
US exact
88.3%
United States

United States benchmark

Exact match
88.3%
Bounded score
96.8%
Parse rate
100.0%
Eligibility flags
956/975

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.

  1. Local income tax100.0%0.3%
  2. Person-level Early Head Start eligibility100.0%3.1%
  3. Person-level Head Start eligibility100.0%1.2%
  4. Person-level WIC eligibility100.0%0.3%
  5. Self-employment tax100.0%2.1%
  6. SSI100.0%2.0%
  7. Person-level Medicare eligibility99.4%11%
  8. Reduced-price school meals eligibility99.0%0.1%
  9. TANF99.0%0.3%
  10. Person-level Medicaid eligibility97.7%30%
  11. Free school meals eligibility97.0%0.7%
  12. Person-level CHIP eligibility94.4%0.2%
  13. Payroll tax92.9%16%
  14. Federal refundable credits90.8%3.7%
  15. SNAP86.0%4.1%
  16. State refundable credits85.7%0.6%
  17. State tax before refundable credits69.8%6.0%
  18. Federal tax before refundable credits58.5%19%
Table view (exact, within 1%, outputs), hardest first
ProgramExactWithin 1%n
Federal tax before refundable credits58.5%69.5%82
State tax before refundable credits69.8%73.3%86
State refundable credits85.7%86.7%98
SNAP86.0%87.1%93
Federal refundable credits90.8%94.9%98
Payroll tax92.9%94.9%99
Person-level CHIP eligibility94.4%94.4%177
Free school meals eligibility97.0%97.0%100
Person-level Medicaid eligibility97.7%97.7%173
Reduced-price school meals eligibility99.0%99.0%100
TANF99.0%99.0%100
Person-level Medicare eligibility99.4%99.4%172
Local income tax100.0%100.0%100
Person-level Early Head Start eligibility100.0%100.0%38
Person-level Head Start eligibility100.0%100.0%38
Person-level WIC eligibility100.0%100.0%177
Self-employment tax100.0%100.0%100
SSI100.0%100.0%97

Hardest casesworst misses on positive references

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 7,772 rows whose legacy threshold score is below 1. That universe contains 7,768 of 7,768 exact-match misses and 4 exact hits. Another 2,027 rows with bounded score below 100 were not selected and have no audit annotation.