Compare AI models

Specifications
Parameters
16B—
Context window
8k32k
Benchmarks
LiveCodeBench
14.3%38.7%
Overview
CompanyMoonshot AIQwen
Release dateFeb 22 2025Jan 28 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Max leads Moonlight on 1 of the 1 benchmark they both report (LiveCodeBench). Qwen2.5-Max shipped 25 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Context windows are 8k (Moonlight) vs 32k (Qwen2.5-Max). Moonlight is open weight, while Qwen2.5-Max is closed.

On LiveCodeBench, Qwen2.5-Max leads at 38.7% vs Moonlight at 14.3%.