Compare AI models

Specifications
Parameters
16B1.5B/4B
Context window
8k8k
Benchmarks
MMLU
70%54.84%
GSM8K
77.4%54.89%
Overview
CompanyMoonshot AIZ.ai
Release dateFeb 22 2025Nov 29 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Moonlight leads GLM-Edge on 2 of the 2 benchmarks they both report (MMLU, GSM8K). GLM-Edge shipped 85 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Moonlight has 16B parameters, while GLM-Edge has 1.5B/4B. Context windows are 8k (Moonlight) vs 8k (GLM-Edge).

On MMLU, Moonlight leads at 70% vs GLM-Edge at 54.84%. On GSM8K, Moonlight leads at 77.4% vs GLM-Edge at 54.89%.