Skip to main content
PolicyBench

DeepSeek V4.1 Flash

DeepSeek · AI alone, no tools
Current board, snapshot 2026-09-29
United States rank
#15 of 45
US exact
87.1%
United States

United States benchmark

Exact match
87.1%
Bounded score
95.1%
Parse rate
100.0%
Eligibility flags
952/975

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. Reduced-price school meals eligibility99.0%0.1%
  7. TANF99.0%0.3%
  8. Free school meals eligibility98.0%0.7%
  9. SSI97.9%2.0%
  10. Person-level CHIP eligibility97.2%0.2%
  11. Person-level Medicaid eligibility96.0%30%
  12. Person-level Medicare eligibility95.3%11%
  13. Federal refundable credits89.8%3.7%
  14. Payroll tax88.9%16%
  15. SNAP84.9%4.1%
  16. State refundable credits83.7%0.6%
  17. State tax before refundable credits65.1%6.0%
  18. Federal tax before refundable credits63.4%19%
Table view (exact, within 1%, outputs), hardest first
ProgramExactWithin 1%n
Federal tax before refundable credits63.4%65.9%82
State tax before refundable credits65.1%67.4%86
State refundable credits83.7%83.7%98
SNAP84.9%89.2%93
Payroll tax88.9%92.9%99
Federal refundable credits89.8%93.9%98
Person-level Medicare eligibility95.3%95.3%172
Person-level Medicaid eligibility96.0%96.0%173
Person-level CHIP eligibility97.2%97.2%177
SSI97.9%99.0%97
Free school meals eligibility98.0%98.0%100
Reduced-price 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
Person-level Head Start eligibility100.0%100.0%38
Person-level WIC eligibility100.0%100.0%177
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: 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.