GLM-5.3-Flash (preview)
Queried as "Ox Alpha", OpenRouter's stealth listing of 2026-08-20; Z.ai identified it as GLM-5.3-Flash after the run. Whether the checkpoint that answered is the one later released under that name has not been stated, so the row keeps its preview qualifier. Its id stays ox-alpha.
United States benchmark
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.
- Local income tax100.0%0.3%
- Person-level Early Head Start eligibility100.0%3.1%
- Person-level Head Start eligibility100.0%1.2%
- Person-level WIC eligibility100.0%0.3%
- Person-level Medicare eligibility99.4%11%
- Reduced-price school meals eligibility99.0%0.1%
- Self-employment tax99.0%2.1%
- TANF99.0%0.3%
- SSI99.0%2.0%
- Free school meals eligibility98.0%0.7%
- Person-level Medicaid eligibility93.8%30%
- Person-level CHIP eligibility92.7%0.2%
- Payroll tax87.0%16%
- Federal refundable credits84.0%3.7%
- SNAP81.4%4.1%
- State refundable credits79.0%0.6%
- Federal tax before refundable credits63.0%19%
- State tax before refundable credits55.0%6.0%
Table view (exact, within 1%, outputs), hardest first
| Program | Exact | Within 1% | n |
|---|---|---|---|
| State tax before refundable credits | 55.0% | 59.0% | 100 |
| Federal tax before refundable credits | 63.0% | 70.0% | 100 |
| State refundable credits | 79.0% | 79.0% | 100 |
| SNAP | 81.4% | 85.6% | 97 |
| Federal refundable credits | 84.0% | 88.0% | 100 |
| Payroll tax | 87.0% | 89.0% | 100 |
| Person-level CHIP eligibility | 92.7% | 92.7% | 177 |
| Person-level Medicaid eligibility | 93.8% | 93.8% | 177 |
| Free school meals eligibility | 98.0% | 98.0% | 100 |
| SSI | 99.0% | 99.0% | 97 |
| Reduced-price school meals eligibility | 99.0% | 99.0% | 100 |
| Self-employment tax | 99.0% | 99.0% | 100 |
| TANF | 99.0% | 99.0% | 100 |
| Person-level Medicare eligibility | 99.4% | 99.4% | 172 |
| Local income tax | 100.0% | 100.0% | 100 |
| Person-level Early Head Start eligibility | 100.0% | 100.0% | 38 |
| Person-level Head Start eligibility | 100.0% | 100.0% | 38 |
| Person-level WIC eligibility | 100.0% | 100.0% | 177 |
Hardest casesworst misses on positive references
- State tax before refundable credits630% offReference $514 · predicted $3,755Inspect household #039
- Federal tax before refundable credits539% offReference $1,346 · predicted $8,596Inspect household #039
- State tax before refundable credits194% offReference $7 · predicted $20Inspect household #023
- SNAP110% offReference $2,428 · predicted $5,100Inspect household #079
- Federal tax before refundable credits102% offReference $3,256 · predicted $6,573Inspect household #086
- State refundable credits100% offReference $25 · predicted $0Inspect household #013
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 8,783 rows whose legacy threshold score is below 1. That universe contains 8,780 of 8,780 exact-match misses and 3 exact hits. Another 1,605 rows with bounded score below 100 were not selected and have no audit annotation.