Compare AI models

Specifications
Parameters
14.3B6B
Context window
32k2k
Benchmarks
HumanEval
34.2%0%
MMLU
62.5%25.2%
GSM8K
61.5%1.5%
Overview
CompanyQwenZ.ai
Release dateMar 28 2024Mar 14 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen1.5-MoE leads ChatGLM-6B on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). ChatGLM-6B shipped 380 days before Qwen1.5-MoE, so benchmark comparisons should account for the intervening progress.

Qwen1.5-MoE has 14.3B parameters, while ChatGLM-6B has 6B. Context windows are 32k (Qwen1.5-MoE) vs 2k (ChatGLM-6B).

On HumanEval, Qwen1.5-MoE leads at 34.2% vs ChatGLM-6B at 0%. On MMLU, Qwen1.5-MoE leads at 62.5% vs ChatGLM-6B at 25.2%. On GSM8K, Qwen1.5-MoE leads at 61.5% vs ChatGLM-6B at 1.5%.