GLM 4.5 Air: benchmark results
Zhipu's lightweight GLM-4.5 Air open MoE (106B total, 12B active). Provider: Zhipu. Released 2025-07-28. Access: Open.
Unified ELO 1584 ± 1, rank #309 of 1392 rated models, from 428 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Galileo Agent - Investment TSQ | 96 | Task Success Quality (%) | 100 |
| SEAL - Fortress | 63.18 | Score | 98.2 |
| Galileo Agent - Banking TSQ | 94 | Task Success Quality (%) | 97.6 |
| Galileo Agent - Insurance TSQ | 94 | Task Success Quality (%) | 97.6 |
| Galileo Agent - Telecom TSQ | 94 | Task Success Quality (%) | 95.2 |
| Evals for Every Language - Language crs | 100 | Average Score (%) | 93.3 |
| FMNB Leaderboard | 99 | Score | 92.7 |
| Evals for Every Language - Language bik | 100 | Average Score (%) | 92.4 |
| Evals for Every Language - Language bew | 100 | Average Score (%) | 91.7 |
| ZeroEval MATH-500 | 98.1 | MATH-500 Score | 90.3 |
| Galileo Agent - Healthcare TSQ | 91 | Task Success Quality (%) | 88.1 |
| Medmarks - HEAD-QA v2 | 88.46 | Score (%) | 87.1 |
Interactive version: theaggregate.ai/model?slug=glm-4-5-air · How It Works · Data refreshed daily, snapshot 2026-09-05.