GLM-5 (Thinking): benchmark results
GLM-5 evaluated with thinking enabled. Provider: Zhipu. Released 2026-02-12. Access: Open.
Unified ELO 1645 ± 1, rank #259 of 1761 rated models, from 67 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA TAU-2 Bench | 98.25 | Accuracy (%) | 98.4 |
| AA Omniscience - Software Engineering (SWE) - Dart | 42 | Accuracy (%) | 96.4 |
| UGI Leaderboard | 54.56 | UGI Score | 95.9 |
| AA Omniscience - Software Engineering (SWE) - Go | 42 | Accuracy (%) | 94 |
| AA Omniscience - Software Engineering (SWE) - C | 67 | Accuracy (%) | 93.7 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 52.73 | Accuracy (%) | 93.5 |
| BenchTable | 74.6 | Total Score (%) | 92.7 |
| AA Omniscience - Software Engineering (SWE) - PHP | 48 | Accuracy (%) | 92.3 |
| AA Omniscience - Software Engineering (SWE) - Java | 31 | Accuracy (%) | 92 |
| UGI - Natural Intelligence | 55.34 | NatInt Score | 92 |
| Kagi LLM Benchmark | 75 | Accuracy (%) | 91.5 |
| UGI - Writing | 55.03 | Writing Score | 90.8 |
Interactive version: theaggregate.ai/model?slug=glm-5-thinking · How It Works · Data refreshed daily, snapshot 2026-09-05.