Compare AI models

Specifications
Parameters
0.5B/1.8B/4B/7B/14B/72B—
Context window
32k—
Benchmarks
HumanEval
71.3%72%
MMLU
75.6%81.5%
GSM8K
82.7%87.6%
Overview
CompanyQwenZ.ai
Release dateFeb 4 2024Jan 16 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GLM-4 leads Qwen1.5 on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). GLM-4 shipped 19 days before Qwen1.5, so benchmark comparisons should account for the intervening progress.

Qwen1.5 is open weight, while GLM-4 is closed.

On HumanEval, GLM-4 leads at 72% vs Qwen1.5 at 71.3%. On MMLU, GLM-4 leads at 81.5% vs Qwen1.5 at 75.6%. On GSM8K, GLM-4 leads at 87.6% vs Qwen1.5 at 82.7%.