Compare AI models

Specifications
Parameters
24B16B
Context window
40k8k
Benchmarks
LiveCodeBench
55.84%14.3%
GPQA Diamond
68.18%35.2%
Overview
CompanyMistralMoonshot AI
Release dateJun 10 2025Feb 22 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Magistral Small leads Moonlight on 2 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). Moonlight shipped 108 days before Magistral Small, so benchmark comparisons should account for the intervening progress.

Magistral Small has 24B parameters, while Moonlight has 16B. Context windows are 40k (Magistral Small) vs 8k (Moonlight).

On LiveCodeBench, Magistral Small leads at 55.84% vs Moonlight at 14.3%. On GPQA Diamond, Magistral Small leads at 68.18% vs Moonlight at 35.2%.