Compare AI models

Specifications
Parameters
14.3B6B
Context window
32k32k
Benchmarks
HumanEval
34.2%9.8%
MMLU
62.5%45.2%
GSM8K
61.5%25.9%
Overview
CompanyQwenZ.ai
Release dateMar 28 2024Jun 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen1.5-MoE leads ChatGLM2-6B on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). ChatGLM2-6B shipped 277 days before Qwen1.5-MoE, so benchmark comparisons should account for the intervening progress.

Qwen1.5-MoE has 14.3B parameters, while ChatGLM2-6B has 6B. Context windows are 32k (Qwen1.5-MoE) vs 32k (ChatGLM2-6B).

On HumanEval, Qwen1.5-MoE leads at 34.2% vs ChatGLM2-6B at 9.8%. On MMLU, Qwen1.5-MoE leads at 62.5% vs ChatGLM2-6B at 45.2%. On GSM8K, Qwen1.5-MoE leads at 61.5% vs ChatGLM2-6B at 25.9%.