Compare AI models

Specifications
Parameters
14.3B9B
Context window
32k128k
Benchmarks
HumanEval
34.2%71.8%
MMLU
62.5%72.4%
GSM8K
61.5%79.6%
Overview
CompanyQwenZ.ai
Release dateMar 28 2024Jun 5 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

GLM-4-9B leads Qwen1.5-MoE on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Qwen1.5-MoE shipped 69 days before GLM-4-9B, so benchmark comparisons should account for the intervening progress.

Qwen1.5-MoE has 14.3B parameters, while GLM-4-9B has 9B. Context windows are 32k (Qwen1.5-MoE) vs 128k (GLM-4-9B).

On HumanEval, GLM-4-9B leads at 71.8% vs Qwen1.5-MoE at 34.2%. On MMLU, GLM-4-9B leads at 72.4% vs Qwen1.5-MoE at 62.5%. On GSM8K, GLM-4-9B leads at 79.6% vs Qwen1.5-MoE at 61.5%.