Compare AI models

Specifications
Parameters
16B6B
Context window
8k32k
Benchmarks
HumanEval
48.1%9.8%
MATH
45.3%6.9%
MMLU
70%45.2%
GSM8K
77.4%25.9%
Overview
CompanyMoonshot AIZ.ai
Release dateFeb 22 2025Jun 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Moonlight leads ChatGLM2-6B on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). ChatGLM2-6B shipped 608 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Moonlight has 16B parameters, while ChatGLM2-6B has 6B. Context windows are 8k (Moonlight) vs 32k (ChatGLM2-6B).

On HumanEval, Moonlight leads at 48.1% vs ChatGLM2-6B at 9.8%. On MATH, Moonlight leads at 45.3% vs ChatGLM2-6B at 6.9%. On MMLU, Moonlight leads at 70% vs ChatGLM2-6B at 45.2%. On GSM8K, Moonlight leads at 77.4% vs ChatGLM2-6B at 25.9%.