GLM-4 Plus: benchmark results

Zhipu's flagship GLM-4-generation API model (128K context), positioned on par with GPT-4o on language understanding and long-text tasks. Provider: Zhipu. Released 2024-08-29. Access: API.

Unified ELO 1666 ± 16, rank #409 of 2656 rated models, from 222 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OpenCompass Language - Dialogue - Chinese (CompassBench 2409)67Score (%)100
OpenCompass Language - Dialogue - English (CompassBench 2409)61.8Score (%)100
OpenCompass Language - Dialogue - English (CompassBench 2411)68Score (%)100
OpenCompass Reasoning - Common Sense - English (CompassBench 2409)58.2Score (%)100
OpenCompass Reasoning - Humanities - Chinese (CompassBench 2409)66.8Score (%)100
OpenCompass Reasoning - Science and Engineering - Chinese (CompassBench 2409)65Score (%)100
OpenCompass Reasoning - Social - Chinese (CompassBench 2409)60.6Score (%)100
OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2409)58.2Score (%)98.3
OpenCompass Reasoning - Common Sense - Chinese (CompassBench 2411)69.2Score (%)97
OpenCompass Reasoning - Humanities - Chinese (CompassBench 2411)70.1Score (%)97
OpenCompass Reasoning - Social - Chinese (CompassBench 2411)67.1Score (%)97
SuperCLUE General (October 2024) - Hard51.09Score95.1

Interactive version: theaggregate.ai/model?slug=glm-4-plus · How It Works · Data refreshed daily, snapshot 2026-09-19.