Compare AI models

Specifications
Parameters
14B6B
Context window
2k32k
Benchmarks
HumanEval
32.3%9.8%
MATH
24.8%6.9%
MMLU
66.3%45.2%
GSM8K
61.3%25.9%
Overview
CompanyQwenZ.ai
Release dateSep 25 2023Jun 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen-14B leads ChatGLM2-6B on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). ChatGLM2-6B shipped 92 days before Qwen-14B, so benchmark comparisons should account for the intervening progress.

Qwen-14B has 14B parameters, while ChatGLM2-6B has 6B. Context windows are 2k (Qwen-14B) vs 32k (ChatGLM2-6B).

On HumanEval, Qwen-14B leads at 32.3% vs ChatGLM2-6B at 9.8%. On MATH, Qwen-14B leads at 24.8% vs ChatGLM2-6B at 6.9%. On MMLU, Qwen-14B leads at 66.3% vs ChatGLM2-6B at 45.2%. On GSM8K, Qwen-14B leads at 61.3% vs ChatGLM2-6B at 25.9%.