Skip to main content
PolicyBench

GLM-5.3

Z.ai · AI alone, no tools
Current board, snapshot 2026-10-14
United States rank
#31 of 50
US exact
81.1%
United States

United States benchmark

Exact match
81.1%
Bounded score
89.6%
Parse rate
96.1%
Eligibility flags
911/975

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. Self-employment tax97.0%2.1%
  2. Local income tax96.9%0.3%
  3. SSI95.9%2.0%
  4. Free school meals eligibility95.0%0.7%
  5. Reduced-price school meals eligibility95.0%0.1%
  6. Person-level WIC eligibility94.9%0.3%
  7. Person-level Medicare eligibility94.8%11%
  8. TANF94.0%0.3%
  9. Person-level CHIP eligibility93.2%0.2%
  10. Person-level Early Head Start eligibility92.1%3.1%
  11. Person-level Head Start eligibility92.1%1.2%
  12. Person-level Medicaid eligibility89.6%30%
  13. Payroll tax84.4%16%
  14. Federal refundable credits81.0%3.7%
  15. SNAP80.6%4.1%
  16. State refundable credits77.8%0.6%
  17. Federal tax before refundable credits54.8%19%
  18. State tax before refundable credits52.7%6.0%
Table view (exact, within 1%, outputs), hardest first
ProgramExactWithin 1%n
State tax before refundable credits52.7%58.1%93
Federal tax before refundable credits54.8%59.5%84
State refundable credits77.8%77.8%99
SNAP80.6%83.9%93
Federal refundable credits81.0%82.0%100
Payroll tax84.4%87.5%96
Person-level Medicaid eligibility89.6%89.6%173
Person-level Early Head Start eligibility92.1%92.1%38
Person-level Head Start eligibility92.1%92.1%38
Person-level CHIP eligibility93.2%93.2%177
TANF94.0%94.0%100
Person-level Medicare eligibility94.8%94.8%172
Person-level WIC eligibility94.9%94.9%177
Free school meals eligibility95.0%95.0%100
Reduced-price school meals eligibility95.0%95.0%100
SSI95.9%96.9%97
Local income tax96.9%96.9%98
Self-employment tax97.0%97.0%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 8,765 rows whose legacy threshold score is below 1. That universe contains 8,761 of 8,761 exact-match misses and 4 exact hits. Another 2,329 rows with bounded score below 100 were not selected and have no audit annotation.