GLM-5 (Thinking): benchmark results
GLM-5 evaluated with thinking enabled. Provider: Zhipu. Released 2026-02-12. Access: Open.
Aggregate Rating 1694 ± 8, rank #378 of 2168 rated models, from 189 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BALSAM - Information Extraction | 53.79 | Overall score (0-100, LLM-judged generation and multiple cho | 100 |
| BALSAM - Logic | 42.41 | Overall score (0-100, LLM-judged generation and multiple cho | 100 |
| BALSAM - Translation/Transliteration | 70.18 | Overall score (0-100, LLM-judged generation and multiple cho | 100 |
| General365 - Branching & Enumeration | 69.1 | Accuracy (%) on the problems labelled Branching & Enumeratio | 100 |
| General365 - Optimal Strategy | 51.7 | Accuracy (%) on the problems labelled Optimal Strategy (a pr | 100 |
| HPR-Bench - Tag Inference - Lifestyle Indicators | 70.98 | Avg@10 Tag Accuracy (%; mean of 10 samples) | 100 |
| Trainee-Bench | 72 | Checkpoint Score (%; mean share of passed checkpoints per sc | 100 |
| Trainee-Bench - Success Rate | 45 | Task Success Rate (%; tasks fully completed) | 100 |
| Tau2-Bench Telecom | 98.25 | Success Rate (%) | 98.6 |
| AA TAU-2 Bench | 98.25 | Accuracy (%) | 98.4 |
| BenchTable - Reasoning | 87.6 | Weighted Score (%) | 97.7 |
| HeaRTS - Statistical Calculation | 94 | Mean task score (x100) over HeaRTS's statistical calculation | 96.7 |
Interactive version: theaggregate.ai/model?slug=glm-5-thinking · How It Works · Data refreshed daily, snapshot 2026-10-20.