Compare AI models

Specifications
Parameters
0.5B/1.5B/7B/57B/72B—
Context window
128k—
Benchmarks
HumanEval
86%72%
MATH
69%47.9%
MMLU
82.3%81.5%
GSM8K
91.1%87.6%
Overview
CompanyQwenZ.ai
Release dateJun 6 2024Jan 16 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen2 leads GLM-4 on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). GLM-4 shipped 142 days before Qwen2, so benchmark comparisons should account for the intervening progress.

Qwen2 is open weight, while GLM-4 is closed.

On HumanEval, Qwen2 leads at 86% vs GLM-4 at 72%. On MATH, Qwen2 leads at 69% vs GLM-4 at 47.9%. On MMLU, Qwen2 leads at 82.3% vs GLM-4 at 81.5%. On GSM8K, Qwen2 leads at 91.1% vs GLM-4 at 87.6%.