GLM-5 (Thinking) — benchmark results
GLM-5 evaluated with thinking enabled. Provider: Zhipu. Released 2026-02-12. Access: Open.
Unified ELO 1781 ± 15, rank #161 of 1776 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA TAU-2 Bench | 98.25 | Accuracy (%) | 98.4 |
| UGI Leaderboard | 54.56 | UGI Score | 96 |
| BenchTable | 74.6 | Total Score (%) | 92.7 |
| UGI - Natural Intelligence | 55.34 | NatInt Score | 92.7 |
| UGI - Writing | 55.03 | Writing Score | 91.5 |
| Kagi LLM Benchmark | 75 | Accuracy (%) | 91.4 |
| AA Terminal-Bench Hard | 43.18 | Accuracy (%) | 90.3 |
| Artificial Analysis Intelligence Index | 39.5 | Intelligence Index | 89.1 |
| AA IFBench | 72.31 | Accuracy (%) | 89 |
| AA Omniscience | 2 | Score | 88.4 |
| AA Omniscience - Software Engineering (SWE) - Dart | 44 | Accuracy (%) | 88.2 |
| AA SciCode | 46.18 | Accuracy (%) | 87.4 |
Interactive version: theaggregate.ai/model?slug=glm-5-thinking · How the rankings work · Data refreshed daily, snapshot 2026-07-22.