Skip to main content
PolicyBench

GLM-5.2

Z.ai · AI alone, no tools
United States rank
#16 of 23
US exact
73.1%
United States

United States benchmark

Exact match
73.1%
Bounded score
84.0%
Parse rate
93.0%
Eligibility flags
894/984

Score by programhardest first

ProgramExactWithin 1%n
Federal tax before refundable credits43.0%43.0%100
State tax before refundable credits45.0%46.0%100
State refundable credits72.0%72.0%100
Payroll tax73.0%80.0%100
SNAP76.0%78.0%100
Federal refundable credits79.0%79.0%100
Person-level Medicaid eligibility85.9%85.9%177
Person-level Medicare eligibility89.3%89.3%177
Person-level Early Head Start eligibility89.5%89.5%38
Person-level Head Start eligibility89.5%89.5%38
SSI90.0%90.0%100
Self-employment tax91.0%91.0%100
Person-level WIC eligibility92.7%92.7%177
Free school meals eligibility93.0%93.0%100
Person-level CHIP eligibility93.2%93.2%177
Local income tax94.0%94.0%100
Reduced-price school meals eligibility94.0%94.0%100
TANF94.0%94.0%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.