Compare AI models

Specifications
Parameters
16B6B
Context window
8k8k
Benchmarks
HumanEval
48.1%58.5%
MATH
45.3%25.7%
MMLU
70%61.4%
GSM8K
77.4%72.3%
Overview
CompanyMoonshot AIZ.ai
Release dateFeb 22 2025Oct 27 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Moonlight leads ChatGLM3-6B on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). ChatGLM3-6B shipped 484 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Moonlight has 16B parameters, while ChatGLM3-6B has 6B. Context windows are 8k (Moonlight) vs 8k (ChatGLM3-6B).

On HumanEval, ChatGLM3-6B leads at 58.5% vs Moonlight at 48.1%. On MATH, Moonlight leads at 45.3% vs ChatGLM3-6B at 25.7%. On MMLU, Moonlight leads at 70% vs ChatGLM3-6B at 61.4%. On GSM8K, Moonlight leads at 77.4% vs ChatGLM3-6B at 72.3%.