GLM-5 — benchmark results
Zhipu's flagship GLM-5 open MoE model (40B active). Provider: Zhipu. Released 2026-02-12. Access: Open.
Unified ELO 1753 ± 12, rank #191 of 1776 rated models, from 295 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AGC-Bench - grapheval_ai_researcher | 1.88 | Dataset z-score | 100 |
| AGC-Bench - schnovel | 2.04 | Dataset z-score | 100 |
| Are LLMs Ready to Assist Physicians? PhysAssis | 69.4 | en_mrs (self-reported) | 100 |
| Ego2World | 183 | Goal Tasks (self-reported) | 100 |
| From 0-Order Selection to 2-Order Judgment | 38.33 | H-Comb (self-reported) | 100 |
| OpenCompass Research - LiveCodeBench v6 | 86.2 | Score (%) | 100 |
| TuRTLe Code Completion (Icarus Verilog) | 83.98 | Aggregated Score (self-reported) | 100 |
| TuRTLe Code Completion (Verilator) | 82.23 | Aggregated Score (self-reported) | 100 |
| AGC-Bench - moh_x | 0.88 | Dataset z-score | 99.4 |
| AGC-Bench - STEM | 0.76 | JRT z-score | 98.8 |
| AGC-Bench - outline_to_story | 1.74 | Dataset z-score | 98.8 |
| Story Theory Bench | 99.6 | Score (%) | 98.5 |
Interactive version: theaggregate.ai/model?slug=glm-5 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.