GLM-5.2 (High) — benchmark results
GLM-5.2 evaluated at the high reasoning-effort setting. Provider: Zhipu. Released 2026-06-13. Access: Open.
Unified ELO 1856 ± 22, rank #101 of 1776 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| WeirdML | 67.31 | Average Score | 85.4 |
| LisanBench | 0.13 | Mean Path Length / Current Maximum | 78.5 |
| OckBench | 78.5 | Accuracy (%) | 68.8 |
| Opus Magnum Bench | 25.39 | Human-normalized score (%) | 61.9 |
| Surface Evolver Bench Pass Rate | 31.25 | Pass Rate (%) | 50 |
| Surface Evolver Bench | 55.62 | Mean Score (%) | 45.5 |
| Epoch AI - Cursorbench | 50.7 | Score | 29.6 |
| CursorBench 3.1 | 51.5 | Score (%) | 24.3 |
| Aikido CVE Rediscovery (pass@3) | 50 | Recall (%) | 2.2 |
| Aikido CVE Rediscovery | 37.2 | Recall (%) | 0 |
Interactive version: theaggregate.ai/model?slug=glm-5-2-high · How the rankings work · Data refreshed daily, snapshot 2026-07-22.