Compare AI models

Specifications
Parameters
6B1.5B/4B
Context window
32k8k
Benchmarks
MMLU
45.2%54.84%
GSM8K
25.9%54.89%
Overview
CompanyZ.aiZ.ai
Release dateJun 25 2023Nov 29 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

GLM-Edge leads ChatGLM2-6B on 2 of the 2 benchmarks they both report (MMLU, GSM8K). ChatGLM2-6B shipped 523 days before GLM-Edge, so benchmark comparisons should account for the intervening progress.

ChatGLM2-6B has 6B parameters, while GLM-Edge has 1.5B/4B. Context windows are 32k (ChatGLM2-6B) vs 8k (GLM-Edge).

On MMLU, GLM-Edge leads at 54.84% vs ChatGLM2-6B at 45.2%. On GSM8K, GLM-Edge leads at 54.89% vs ChatGLM2-6B at 25.9%.