Compare AI models

Specifications
Parameters
110B—
Context window
32k128k
Benchmarks
HumanEval
52.4%75.7%
MATH
49.6%57.9%
MMLU
80.4%81.9%
GSM8K
85.4%90.9%
Overview
CompanyQwenZ.ai
Release dateApr 25 2024Jun 5 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GLM-4-Air leads Qwen1.5-110B on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). Qwen1.5-110B shipped 41 days before GLM-4-Air, so benchmark comparisons should account for the intervening progress.

Context windows are 32k (Qwen1.5-110B) vs 128k (GLM-4-Air). Qwen1.5-110B is open weight, while GLM-4-Air is closed.

On HumanEval, GLM-4-Air leads at 75.7% vs Qwen1.5-110B at 52.4%. On MATH, GLM-4-Air leads at 57.9% vs Qwen1.5-110B at 49.6%. On MMLU, GLM-4-Air leads at 81.9% vs Qwen1.5-110B at 80.4%. On GSM8K, GLM-4-Air leads at 90.9% vs Qwen1.5-110B at 85.4%.