GLM-5.2 (Max) — benchmark results

GLM-5.2 evaluated at the max reasoning-effort setting. Provider: Zhipu. Released 2026-06-13. Access: Open.

Unified ELO 1925 ± 25, rank #52 of 1776 rated models, from 65 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA TAU-2 Bench99.12Accuracy (%)99.9
WebDev Arena1593.25Arena Score98.7
AA Humanity's Last Exam40.13Accuracy (%)97.2
Artificial Analysis Intelligence Index51.09Intelligence Index97
Chatbot Arena (Code)1592Elo97
AA CritPt20.86Accuracy (%)96.6
AA Long Context Reasoning71.33Accuracy (%)95.4
AA Terminal-Bench Hard50.76Accuracy (%)95
AA SciCode50.46Accuracy (%)93.9
AA GPQA Diamond89.49Accuracy (%)92.5
LLM2014 Logic 2026-0671.16Median Score92.5
GDPval-AA1514Elo92.3

Interactive version: theaggregate.ai/model?slug=glm-5-2-max · How the rankings work · Data refreshed daily, snapshot 2026-07-22.