Compare AI models

Specifications
Parameters
16B7B
Context window
8k64k
Benchmarks
HumanEval
48.1%83.5%
Overview
CompanyMoonshot AIQwen
Release dateFeb 22 2025Apr 16 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

CodeQwen1.5 leads Moonlight on 1 of the 1 benchmark they both report (HumanEval). CodeQwen1.5 shipped 312 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Moonlight has 16B parameters, while CodeQwen1.5 has 7B. Context windows are 8k (Moonlight) vs 64k (CodeQwen1.5).

On HumanEval, CodeQwen1.5 leads at 83.5% vs Moonlight at 48.1%.