GLM-4.5 (Thinking): benchmark results
GLM-4.5 evaluated with thinking enabled. Provider: Zhipu. Released 2025-07-28. Access: Open.
Unified ELO 1606 ± 1, rank #409 of 1761 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM2014 Code 2025-09 - TypeScript | 8.89 | Score | 92.1 |
| LLM2014 Code 2025-09 - C++ | 5.62 | Score | 84.2 |
| LLM2014 Code 2025-09 - Python | 7.15 | Score | 78.9 |
| BenchTable | 60.8 | Total Score (%) | 71.7 |
| LLM2014 Logic 2025-08 | 42.18 | Median Score | 59.1 |
| LLM2014 Code 2025-09 - C# | 6.94 | Score | 57.9 |
| LLM2014 Code 2025-09 | 37.21 | Median Score | 55.6 |
| LisanBench | 0.03 | Mean Path Length / Current Maximum | 53.6 |
| LLM2014 Logic 2025-09 | 37.61 | Median Score | 51.1 |
| Epoch AI - Algotune | 1.52 | Score | 50 |
| VitaBench 2.0 | 36.9 | Avg@4 Full Context (self-reported) | 47.1 |
| WeirdML | 40.58 | Average Score | 36.9 |
Interactive version: theaggregate.ai/model?slug=glm-4-5-thinking · How It Works · Data refreshed daily, snapshot 2026-09-05.