Compare AI models

Specifications
Parameters
16B32B
Context window
8k128k
Benchmarks
LiveCodeBench
14.3%63.4%
Overview
CompanyMoonshot AIQwen
Release dateFeb 22 2025Mar 6 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

QwQ-32B leads Moonlight on 1 of the 1 benchmark they both report (LiveCodeBench). Moonlight shipped 12 days before QwQ-32B, so benchmark comparisons should account for the intervening progress.

Moonlight has 16B parameters, while QwQ-32B has 32B. Context windows are 8k (Moonlight) vs 128k (QwQ-32B).

On LiveCodeBench, QwQ-32B leads at 63.4% vs Moonlight at 14.3%.