GLM-5.2 (Non-reasoning): benchmark results
GLM-5.2 evaluated with reasoning disabled. Provider: Zhipu. Released 2026-06-13. Access: Open.
Aggregate Rating 1690 ± 19, rank #391 of 2168 rated models, from 65 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Writing | 61.21 | Writing Score | 92.3 |
| UGI - Natural Intelligence | 54.88 | NatInt Score | 90.2 |
| AA-Omniscience Hallucination Rate | 33.68 | Hallucination Rate (%) | 89.6 |
| DiagChain - Clean Evidence | 28.27 | Correctly reconstructed reference steps (%; share of the 283 | 80 |
| AA-Omniscience Index - Software Engineering (SWE) - PHP | 28 | Omniscience Index | 77.8 |
| AA-Omniscience Index - Humanities & Social Sciences | -1 | Omniscience Index | 75.5 |
| AA-Omniscience Index - Software Engineering (SWE) - Dart | 6 | Omniscience Index | 75.2 |
| AA-Omniscience Index - Software Engineering (SWE) - Go | 16 | Omniscience Index | 74.8 |
| AA-Omniscience Index - Science, Engineering & Mathematics | -0.3 | Omniscience Index | 74.4 |
| AA-Omniscience Index - Health | -12.7 | Omniscience Index | 74.1 |
| RideWay | 62 | Efficiency Utility (success-gated, held-out custom14 mean ov | 73.9 |
| RideWay - Success Rate | 78.7 | Success rate (%; share of trials passing the three-judge LLM | 73.9 |
Interactive version: theaggregate.ai/model?slug=glm-5-2-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.